{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6febc445-889c-4db1-b014-6a346ab9a49f",
   "metadata": {
    "tags": [
     "remove_cell"
    ]
   },
   "outputs": [],
   "source": [
    "# Setting up the Colab environment. DO NOT EDIT!\n",
    "import os\n",
    "#import warnings\n",
    "#warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "try:\n",
    "    import otter, pingouin\n",
    "\n",
    "except ImportError:\n",
    "    ! pip install -q otter-grader==4.0.0, pingouin\n",
    "    import otter\n",
    "\n",
    "if not os.path.exists('walkthrough-tests'):\n",
    "    zip_files = [f for f in os.listdir() if f.endswith('.zip')]\n",
    "    assert len(zip_files)>0, 'Could not find any zip files!'\n",
    "    assert len(zip_files)==1, 'Found multiple zip files!'\n",
    "    ! unzip {zip_files[0]}\n",
    "\n",
    "grader = otter.Notebook(colab=True,\n",
    "                        tests_dir = 'walkthrough-tests')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cea3b0b0",
   "metadata": {},
   "source": [
    "# Walkthrough"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71197956",
   "metadata": {},
   "source": [
    "## Learning Objectives\n",
    "At the end of this learning activity you will be able to:\n",
    " - Practice using `pg.normality` and `pg.qqplot` to assess normality.\n",
    " - Practice using `pg.linear_regression` to perform multiple regression.\n",
    " - Interpret the results of linear regression such as the coefficient, p-value, R^2, and confidence intervals.\n",
    " - Describe a _residual_ and how to interpret it.\n",
    " - Relate the _dummy variable trap_ and how to avoid it during regression.\n",
    " - Describe _overfitting_ and how to avoid it."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "230f0ff0",
   "metadata": {},
   "source": [
    "As we discussed with Dr. Devlin in the introduction video, this week and next we are going to look at HIV neurocognitive impairment data from a cohort here at Drexel.\n",
    "Each person was given a full-scale neuropsychological exam and the resulting values were aggregated and normalized into Z-scores based on demographically matched healthy individuals.\n",
    "\n",
    "In this walkthrough we will explore the effects of antiretroviral medications on neurological impairment.\n",
    "In our cohort, we have two major drug regimens, d4T (Stavudine) and the newer Emtricitabine/tenofovir (Truvada).\n",
    "The older Stavudine is suspected to have neurotoxic effects that are not found in the newer Truvada.\n",
    "We will use inferential statistics to understand this effect."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a0a08b85-58d9-4963-828b-8b515b8470f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import pingouin as pg\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2d3c415d-aff6-401d-9ffd-61abe1112897",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sex</th>\n",
       "      <th>age</th>\n",
       "      <th>education</th>\n",
       "      <th>race</th>\n",
       "      <th>processing_domain_z</th>\n",
       "      <th>exec_domain_z</th>\n",
       "      <th>language_domain_z</th>\n",
       "      <th>visuospatial_domain_z</th>\n",
       "      <th>learningmemory_domain_z</th>\n",
       "      <th>motor_domain_z</th>\n",
       "      <th>ART</th>\n",
       "      <th>YearsSeropositive</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>male</td>\n",
       "      <td>62</td>\n",
       "      <td>10.0</td>\n",
       "      <td>AA</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.151646</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.152131</td>\n",
       "      <td>-1.364306</td>\n",
       "      <td>Stavudine</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>male</td>\n",
       "      <td>56</td>\n",
       "      <td>10.0</td>\n",
       "      <td>AA</td>\n",
       "      <td>-0.5</td>\n",
       "      <td>1.2</td>\n",
       "      <td>-0.255505</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>-0.086376</td>\n",
       "      <td>-0.348600</td>\n",
       "      <td>Truvada</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>female</td>\n",
       "      <td>51</td>\n",
       "      <td>10.0</td>\n",
       "      <td>AA</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.902004</td>\n",
       "      <td>-0.4</td>\n",
       "      <td>-1.139892</td>\n",
       "      <td>0.112215</td>\n",
       "      <td>Stavudine</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>female</td>\n",
       "      <td>47</td>\n",
       "      <td>12.0</td>\n",
       "      <td>AA</td>\n",
       "      <td>-0.6</td>\n",
       "      <td>-1.2</td>\n",
       "      <td>-0.119866</td>\n",
       "      <td>-2.1</td>\n",
       "      <td>0.803619</td>\n",
       "      <td>-2.276768</td>\n",
       "      <td>Truvada</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>male</td>\n",
       "      <td>46</td>\n",
       "      <td>13.0</td>\n",
       "      <td>AA</td>\n",
       "      <td>-0.4</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.079129</td>\n",
       "      <td>-1.3</td>\n",
       "      <td>-0.533607</td>\n",
       "      <td>-0.330541</td>\n",
       "      <td>Truvada</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      sex  age  education race  processing_domain_z  exec_domain_z  \\\n",
       "0    male   62       10.0   AA                  0.5            0.6   \n",
       "1    male   56       10.0   AA                 -0.5            1.2   \n",
       "2  female   51       10.0   AA                  0.5            0.1   \n",
       "3  female   47       12.0   AA                 -0.6           -1.2   \n",
       "4    male   46       13.0   AA                 -0.4            1.3   \n",
       "\n",
       "   language_domain_z  visuospatial_domain_z  learningmemory_domain_z  \\\n",
       "0           0.151646                   -1.0                -1.152131   \n",
       "1          -0.255505                   -2.0                -0.086376   \n",
       "2           0.902004                   -0.4                -1.139892   \n",
       "3          -0.119866                   -2.1                 0.803619   \n",
       "4           0.079129                   -1.3                -0.533607   \n",
       "\n",
       "   motor_domain_z        ART  YearsSeropositive  \n",
       "0       -1.364306  Stavudine                 13  \n",
       "1       -0.348600    Truvada                 19  \n",
       "2        0.112215  Stavudine                  9  \n",
       "3       -2.276768    Truvada                 24  \n",
       "4       -0.330541    Truvada                 14  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('hiv_neuro_data.csv')\n",
    "data['education'] = data['education'].astype(float)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac31172e-1108-4f2c-a322-07e1f91d0942",
   "metadata": {},
   "source": [
    "Before we start, we need to talk about assumptions.\n",
    "\n",
    "Basic linear regression has a number assumptions baked into itself:\n",
    " - **Linearity**: The relationship between the independent variables (predictors) and the dependent variable (outcome) is linear. This means that changes in the predictors lead to proportional changes in the dependent variable.\n",
    " - **The relationship between the independent variables and the dependent variable is additive**: The effect of changes in an independent variable X on the dependent variable Y is consistent, regardless of the values of other independent variables. This assumption might not hold if there are interaction effects between independent variables that affect the dependent variable.\n",
    " - **Independence**: Observations are independent of each other. This means that the observations do not influence each other, an assumption that is particularly important in time-series data where time-related dependencies can violate this assumption.\n",
    " - **Homoscedasticity**: The variance of error terms (residuals) is constant across all levels of the independent variables. In other words, as the predictor variable increases, the spread (variance) of the residuals remains constant. This is evaluated at the **end** of the fit.\n",
    " - **Normal Distribution of Errors**: The residuals (errors) of the model are normally distributed. This assumption is especially important for hypothesis testing (e.g., t-tests of coefficients) and confidence interval construction. It's worth noting that for large sample sizes, the Central Limit Theorem helps mitigate the violation of this assumption. This is evaluated at the **end** of the fit.\n",
    " - **Minimal Multicollinearity**: The independent variables need to be independent of each other. Multicollinearity doesn't affect the fit of the model as much as it affects the coefficients' estimates, making them unstable and difficult to interpret.\n",
    " - **No perfect multicollinearity**: Also called the _dummy variable trap_. It states that none of the independent variables should be a perfect linear function of other independent variables. We'll talk more about this when we run into it.\n",
    "\n",
    "Biology itself is highly non-linear.\n",
    "That doesn't mean we can't use linear assumptions to explore biological questions, it just means that we need to be mindful when interpretting the results."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6ab9af5-a5ea-451c-b267-fcc0b0b1afd7",
   "metadata": {},
   "source": [
    "## Exploration"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e1954ae-3cb3-4167-8705-e9123c1e9d40",
   "metadata": {},
   "source": [
    "Let's start by plotting the each variable against EDZ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d8dd6aa8-655e-4d6b-a977-1e6d4ed91181",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (age_ax, edu_ax, ys_ax) = plt.subplots(1,3, sharey=True, figsize = (15, 5))\n",
    "\n",
    "sns.regplot(data = data,\n",
    "            x = 'age',\n",
    "            y = 'exec_domain_z',\n",
    "            ax=age_ax)\n",
    "\n",
    "sns.regplot(data = data,\n",
    "            x = 'education',\n",
    "            y = 'exec_domain_z',\n",
    "            ax=edu_ax)\n",
    "\n",
    "sns.regplot(data = data,\n",
    "            x = 'YearsSeropositive',\n",
    "            y = 'exec_domain_z',\n",
    "            ax=ys_ax)\n",
    "\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2c4b2076-e3e1-484e-bd41-31a07f419162",
   "metadata": {
    "deletable": false,
    "editable": false
   },
   "source": [
    "### Q1: By inspection, which variable is most correlated?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e601810-0c65-4d8f-86d6-aa26184e1971",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "|               |    |\n",
    "| --------------|----|\n",
    "| Points        | 5  |\n",
    "| Public Checks | 3  |\n",
    "\n",
    "_Points:_ 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "016c7dda-c8f7-43bd-b956-9eb418126bcc",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [],
   "source": [
    "# Answer: age, education, YearsSeropositive\n",
    "q1_most_correlated = 'YearsSeropositive' # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a31d129f",
   "metadata": {
    "deletable": false,
    "editable": false
   },
   "outputs": [],
   "source": [
    "grader.check(\"q1_initial_correlation\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a11fb13c-1794-4fad-8586-96727cd1ca88",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='ART', ylabel='exec_domain_z'>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABNAAAAHACAYAAABj3JXZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAADC7ElEQVR4nOzde3wV9Z0//tfMOTk5BBICISRAAtFESkALNm5XUQTaWsDuVrt2e9FE3V4EQ1taurS1Wi9brHVZtW6VS3fb0gUv7a77s93HV826FVBKXbdpwRruR0MIkBATSALh5Fxmfn+czOFcZubMzJlzfz0fD1ozZ2bOwHw+8/7MJ+fMS5BlWQYRERERERERERGpEjN9AERERERERERERNmME2hEREREREREREQ6OIFGRERERERERESkgxNoREREREREREREOjiBRkREREREREREpIMTaERERERERERERDo4gUZERERERERERKSDE2hEREREREREREQ6nJk+gHSSJAknT55EaWkpBEHI9OEQEeU8WZYxPDyM6dOnQxT5OxnWGSIie7HOxGOtISKyj5k6U1ATaCdPnkRtbW2mD4OIKO8cP34cNTU1mT6MjGOdISJKDdaZi1hriIjsZ6TOFNQEWmlpKYDQP0xZWVmGj4aIKPcNDQ2htrY2fH0tdKwzRET2Yp2Jx1pDRGQfM3WmoCbQlI84l5WVsdgQEdmIXyEJYZ0hIkoN1pmLWGuIiOxnpM7wQQJEREREREREREQ6OIFGRERERERERESkgxNoREREREREREREOjiBRkREREREREREpIMTaERERERERERERDo4gUZERERERERERKSDE2hEREREREREREQ6OIFGRERERERERESkgxNoREREREREREREOpyZPgAiIiIiIiKiTAgGgzhw4ADOnDmDSZMmobGxEQ6HI9OHRURZiBNoREREREREVHDefPNNbN26FadPnw4vmzp1Ku68805cffXVGTwyIspGnEDLcm0dPdi44ygO957D7KoJWFg/BXs874d/bl3agGXzqjN9mERERJbE1rlk6pqd+yIiovyk1IqDpwbhON+H66Zdjke+cQNmzpyJrq4uvPDCC9iwYQPWrVvHSTTKWRwTpYYgy7Kc6YNIl6GhIUycOBGDg4MoKyvL9OEk1NbRg5Xb2nXXEQRgc3MTOwMRZUSuXVdTjf8e5qjVOat1zc59EVH24HU1Hv9NrDNSKyRJwg9/+EMcP34cTz31FL/OSTmHYyJzzFxT+Qm0DBgdHUV3d3fC9R5/pTPhOrIMPPFKBxrc5204MmNqampQXFyctvcjIqL0M1qrkqFW56zWNTv3lQjrIBFRbtq442jcMlkGNu70hCcWRFHELbfcgnvuuQcHDhzA5Zdfnu7DpCyUjnGRXTgmSh1OoGVAd3c31q1bl3C9Iw23A2JRwvUO9w4b2p9dNmzYgPr6+rS9HxERpZ/RWpUMrTpnpa7Zua9EWAeJiHLT4d5zqsuP9A5H/Txz5kwAwJkzZ1J+TJQb0jEusgvHRKnDCbQMqKmpwYYNGxKu1/piJw71eROuN7uqFBtWhvbX3d2NJ598EmvWrEFNTU3Sx6omVfslIqLsYbRWJUOrzkXWNcBYbTO6LzuwDhIR5abZVROwr3swbvllVaVRP3d1dQEAJk2alJbjouyXjnGRERwTZRYn0DKguLjY0Czt2uXjsWp7O/SeUicIwNrl81BfH/1d5pqamoKaCSYiInsZrVXJUKtzWnUN0K9tZvdFRESFp3Vpg2qtWL3kYm2RJAkvvPACqqqq0NjYmIGjpGyUjnGRGRwTZYaY6QMgbcvmVWNzcxPm15ajxOXA/Npy3L24PurnLc1N+DgfBEhERDlIrc5ZrWt27ouIiPJTZK0odgAl3j7c4H4Ps5yDuHDhAg4dOoQf/vCHaG9vxx133MEAAcpJHBOlDj+BluWWzatmUgYREeUtO+scayYRESUSWSvefPNNbN26Fffc81r49aqqKqxbtw5XX311pg6RKGkcE6UGJ9CIiIiIiIio4Fx99dX4i7/4Cxw4cABnzpzBpEmT0NjYyE+eEZEqTqARERERERFRQXI4HLj88sszfRhElAP4DDQiIiIiIiIiIiIdnEAjIiIiIiIiIiLSwQk0IiIiIiIiIiIiHZxAIyIiIiIiIiIi0sEJNCIiIiIiIiIiIh2cQCMiIiIiIiIiItLBCTQiIiIiIiIiIiIdnEAjIiIiIiIiIiLSwQk0IiIiIiIiIiIiHZxAIyIiIiIiIiIi0sEJNCIiIiIiIiIiIh2cQCMiIiIiIiIiItLBCTQiIiIiIiIiIiIdnEAjIiIiIiIiIiLSwQk0IiIiIiIiIiIiHc5MHwARRQsGgzhw4ADOnDmDSZMmobGxEQ6HI9OHRURERERERFSwOIFGlEXefPNNbN26FadPnw4vmzp1Ku68805cffXVGTwyIiIiIiIiosLFCTSiLPHmm29iw4YNaGpqwje+8Q0cHhmHH//2MN5+fwSvP38Qd5/04at/c32mD5OIyJK2jh5s3HEUh3vPYXbVBLQubcCyedWZPiwiIiKijODYKPfwGWhEWSAYDGLr1q1oamrCd77zHXQGJuKrv/wzDr8/igAcGHFX4rH/HcLLfz6Z6UMlIjKtraMHK7e1Y1/3IC74g9jXPYhV29vR1tGT6UMjIiIiSjuOjXITJ9CIssCBAwdw+vRp3HLLLRBFERt3HI1fSRDwxCsd6T84IqIkqV3TZBnYuNOTgaMhIiIiyiyOjXITJ9CIssCZM2cAADNnzgQAHO49p7resbO+tB0TEZFdtK5pR3qH03wkRERERJnHsVFu4gQaURaYNGkSAKCrqwsAMLtqgup6s8pdaTsmIiK7aF3TLqsqTfOREBEREWUex0a5KWcm0B555BH8xV/8BUpLSzF16lTcfPPNOHToUKYPi8gWjY2NmDp1Kl544QVIkoTWpQ0QhJiVZBlrl8/LyPERESVD7ZomCMDqJfWZOSAiIiKiDOLYKDflzATarl27sHr1arz55pt49dVXEQgE8PGPfxznz5/P9KGlTVtHD256ajcav/cKbnpqNx8wmEccDgfuvPNOtLe344c//CHqnIP4589cgQ9MKYYTQZR4+/D3V5dh+RXTM32oRESmLZtXjc3NTZhfW44SlwPza8uxpbkJH2fSFBHHd0REBcjK2KitowetL3Zib8PtaH2xk/UiA5yZPgCjXnnllaiff/7zn2Pq1Klob2/H9ddfn6GjSh8lpUOhpHRsbm5i1G2euPrqq7Fu3Tps3boV99xzDwBgPIAbqqpwxx134Oqrr87sARIRJWHZvGrWK6IYHN8RERUuM2OjqHohFuFQn5f1IgNyZgIt1uDgIABg8uTJmuuMjo5idHQ0/PPQ0FDKjytV9FI62GHyx9VXX42/+Iu/wIEDB3DmzBlMmjQJjY2NcDgcmT40IlKRT3WGKFeNjo6iu7s704dhyeOvdMYtk2XgiVc60OBO/lsWNTU1KC4uTno/lFmsNUTE+YDskJMTaLIsY+3atbjuuutw+eWXa673yCOP4KGHHkrjkaUOUzoKh8Ph0G3XRJQ98qnOEOWq7u5urFu3LtOHYcmRhtsBsShu+eHeYVv+Ths2bEB9PZ+nk+tYa4iI8wHZIScn0L7yla/g7bffxu7du3XXu+eee7B27drwz0NDQ6itrU314aXE7KoJ2Nc9GLecKR1ERJmTT3WGKFfV1NRgw4YNmT4MAKHJvCeffBJr1qxBTU1NwvVbX+zEoT5v3PLZVaXYsDL5v5ORY6Dsx1pDRJwPyA45N4H21a9+Fb/5zW/w+uuvJxwUFBcX583H1luXNmDV9nbI8sVlTOkgIsqsfKozRLmquLg46z5lVVNTY+iY1i4frzq+W7t8Hurr+ZUcCmGtISLOB2SHnEnhlGUZX/nKV/Cf//mfeO2113DJJZdk+pDSiglmRERERPmF4zsiIjJCqRdzKt0QJT/mVLpZLzIgZz6Btnr1ajz77LP49a9/jdLSUvT0hCJbJ06ciHHjxmX46NIjUUpHW0cPHn+lE0fGYm3XLh/PBwoSEVHBauvowcYdR3G49xxmV01A69IG1kXKOkyoJSIiI5bNq0aD+zzWrVuHDSs3mP6kMsdFycuZT6Bt2rQJg4ODWLJkCaZNmxb+88tf/jLTh5YVlFjbQ31eSBGxtm0dPZk+NCIiorRT6uK+7kFc8Aexr3uQdZGIiIgKEsdF9siZCTRZllX/3HnnnZk+tKygF2tLRERUaFgXiYiIiEI4LrJHzkygkT7G2hIREV3EukhEREQUwnGRPTiBlidmV01QXc5YWyIiKkSsi0REREQhHBfZgxNoeaJ1aQMEIXoZY22JiKhQsS4SERERhXBcZA9OoOUJxtoSERFdpNTF+bXlKHE5ML+2nHWRiIiIChLHRfZwZvoAyD7JxtoSERHlk2XzqhnPTkRERASOi+zAT6ARERERERERERHp4AQaERERERERERGRDk6gERERERERERER6eAEGhERERERERERkQ5OoBEREREREREREelgCmcBaOvowcYdR3G49xxmV01A69IGpm8QERERmcQxFRERWcH6kR/4CbQ819bRg5Xb2rGvexAX/EHs6x7Equ3taOvoyfShEREREeUMjqmIiMgK1o/8wQm0PLdxx9G4ZbIMbNzpycDREBEREeUmjqmIiMgK1o/8wQm0PHe495zq8iO9w2k+EiIiIqLcxTEVERFZwfqRPziBludmV01QXX5ZVWmaj4SIiIgod3FMRUREVrB+5A9OoOW51qUNEIToZYIArF5Sn5kDIiIiIspBHFMREZEVrB/5gxNoeW7ZvGpsbm7C/NpylLgcmF9bji3NTfg4Ez+IiIiIDOOYioiIrGD9yB/OTB8AmdfW0YNHXjqAYwMjEADMnFyCe25s1IzBXTavOu41xuiml/LvfeDUMByigKAko3FaKf/diSivma01idbf3TmML/7njnD9mzKhGCUuB3qHRlnLKC3UxlRElL14z0NZRZYhy2P/D+vt8+z4WWh9sRNdZ48k3I59wF78BFqOUSJwO/tHIMuAJAOd/SNYtc14DC5jdNMr8t/bF5RwwR+ELyjx352I8prZWpNo/bPjZ+GBV09E1b/Tw6Po7B9hLSMioji856FsodYWV25rt9Q+d3cO490ZH8OhPm/C7dgH7MdPoOUYtQhcAJARisF9fHni2WS9GF3ORhs3OjqK7u7uhOs9/kqn5muyDDzxSgca3OdtPDJjampqUFxcnPb3JaLCYLbW6K3/+PJq9FTMT/ierGVERLnF6HjaCrUxeCbH3mZwnJ5ftO7hYxkZxzy7t9/wdrzvtx8n0HKMVgQuoMTgJu4IjNG1R3d3N9atW5dwvSMNtwNikebrh3uHDe3Hbhs2bEB9PR9cSUSpYbbW6K9fDa+r3ND7spYREeUOo+NpK7TG4Jkae5vBcXp+0buHj5VoHHPszKjh7Xjfbz9OoOWY2VUTsK97UPU1ozG4WvtgjK45NTU12LBhQ8L1Wl/sxKE+r+brs6tKsWFl9H66u7vx5JNPYs2aNaipqUn6WNWkar9ERID5WpNofbfvLEbclQnfl7WMiCh3GB1PW6E1Bs/U2NuMbDgGso/ePXysROOYWZOKVdu12na877cfJ9ByTOvSBqzc1h63XIASg5v448itSxuwant76AGGyvaM0TWtuLjY0G+G1i4fH/fvrRAEYO3yeaivV//kYE1NDX/7REQ5yWyt0V//PKr79+HdGR/TfU/WMiKi3GJ0PG2F2hicY2/KBLUxDhC6h49cZGQcc+uCCjzw392hlRNsx/t++3ECLUdEpmfUVZTgvC+A94d9kBHqeLMqSsKdz0gqx6zJJegaGAEQSvH87o2NjNFNESW2eONODw6cGoJzLIVzzrQyrF5Sz393IspLkde+I73DuKyqVPeap7e+x+NB+fljeOiGGXjqzT70DfsAAE5RwOTxRTg3GtTcP9OnKBWstiu2R6L0MVqH2jp68PgrnTjScDtaX+zE2uXjw/2SfZbsoNUWleeYa7VPtfZ3XV0pLj35W7gW/BW6Bv2646tl86qx6vp6/GJPJ0b8QZQUOXDHtXW8/0wCJ9BygJKeoejsH4marZbHlq3a3o7PfnBy6Df0Yx/rVJI2Njc3Ydm86rh9AcCxgRGofDiKbMTYeyIqRGavfUbWVybPACAgyeg75wvXuFixNS+2JhJZodeuGtzWtmN7JEqNRHUlql+KRTjU5w33SwDss2Qbrbao1Za0asaDH5uB8vPHsOHmuoSflmzr6MGmXZ7wzyP+IDbv8mBBbTnbsEVipg+AElNNz1BZT5aBFzvOqC7fuNOjva+I14mIiLKVXvKUGtY8SgWr7YrtkSj76PVL9lnKJK3295zKWMjsPtiGreMn0HKAmdQOb0D9s2RK0gaTOIiIKFeZSZ4CWPMoNRIlxlrbjogyQa9fqj2/WHmNKNW02mbnmVHMTXIfbMPW8RNoOWB21QTD67qdgupyJWlDa19M4iAiomw3a1Kx6nK9ZE8z6xMZYbVdsT0SZR+9fsk+S5mk1f7qNMZCZvbBNmwdJ9ByQOvShsiQDQCh4IBYggB8at4kxP66JDJpQ3VfTOIgIqIccOuCClM1jDWPUsFqu2J7JMo+ev2SfZYySav93bqgIul9sA1bxwm0HKCkdsyvLUeJy4H5teXY0tKELS0xy5qb8KUPT8WlJ3+LOZXuqOVK0obqviJeJyIiylbX1ZWaqmGseZQKVtsV2yNR9lH65ZxKN0TJjzmV7nC/ZJ+lTNJqf9fWGf/0GNuw/fgMtCwXG137xGcXRCVmxKZneDyehKkcTIQkSi1GnhNpS7Z/pCLZk8gsq+0qdru2jh7c9NRu1guiDFo2rxoN7vNYt24dNqzcgPr66oT3YETpoFZrPJ7oAIBE4yqOg+zFT6BlMSW6dl/3IC74g+Ho2raOnkwfGhFpYL8l0sb+QXQR+wNRdmLfpFzBtpp+/ARaFtOLneUsMpE9RkdH0d3dbdv+Hn+lM26ZLANPvNKBBvd5295HUVNTg+Ji4w8TJcok1jWii9gfiLIT+yblCrbV9OMEWhZj7CxR6nV3d2PdunW27e9Iw+2AWBS3/HDvsK3vo9iwYYPm17WJsg3rGtFF7A9E2Yl9k3IF22r6cQIti82umoB93YNxyxk7S2SfmpoabNiwwbb9tb7YiUN93rjls6tKsWGlfe+jqKmpsX2fRKnCukZ0EfsDUXZi36RcwbaafpxAy2KtSxuwans7ZPniMsbOEtmruLjY1k9wrV0+Pr7fAli7fB7q6/lRaipsrGtEF7E/EGUn9k3KFWyr6ccJtCy2bF41Vl1fj1/s6cSIPwhBCN2If+XZP8EflCAIwMzJJbjnxsaE33FmKiBReihx0Rt3enDo1CDE4V5855NXMi6aCPF1raTIgTuurTPUP86On4XWFzvRdfaIah2LrHNVZaHnAvYOjbLmkW3sHktF1osjvcO4rKoUq5fUs14QZcDuzmEcnPlJfOLnh/CB6jKsur4ee97tZ98k28TWkIX1U7DH8z4OnBqGQxQQlGQ0Tis1VVti68jU0tD4Z83zezn+SRFOoGWxto4ebNp1MaZWlgEZgC8ohX/u7B/Bqm3t2NzSpNk5lHQOhZLOsblZexsisk6Ji/Z4PFi37l9xbd31mT4koqwQW9dG/EFs3uXBgtpy3Xq0u3MY7874GDD29ejYOhZb5zr7R8L/zZpHdkjVWEqpF0SUOW0dPXjg1ROAuxIIyNjXPYi3TwyybpBt1GpI1FcvgxeXm60tSh3hPX96iJk+ANKmlqqhRkYoacPMfpR0DiIionSxWo+e3duvu12iesmaR8niWIoof7F/U6oZva8HrLc9tuP04ARaFtNK1VCjl7TBdA4iIsoGVuvRsTOjutsZqZeseZQMjqWI8hf7N6Wamft6wFrbYztOD06gZbHZVRMMr6uXtKG1H6ZzEBFROlmtR7MmFetuZ6ResuZRMjiWIspf7N+Uambu6wFrbY/tOD04gZbFWpc2QBASrydAP2lDbT9M5yAionSzWo9uXVCBqIipmO0S1UvWPEoWx1JE+at1aQNiSwj7N9nJ6H09YL3tsU6lByfQspiSqjG/thwup4gSlwNOUYDLKUIAIApAXUUJtrQ06abCRO6nxOXA/NpybGnW34aIiMhuVuvRdXWluPTkbzGn0q26Xex+6ypKUDdlPGse2YZjKaL8tWxeNR68YQZKLpyG2ymwf5Pt1GrI3Uvqo+7zi51iUm2PdSo9mMKZ5ZTEDCXydt70snDkrfI95x+8dCAcVfs3jRc/uvnoywexdU8nLviDEBAKGwCAt4+fxQ9eOgA5Yv9ERETpkCh1UIl5j4x1v2SyCwDw9M11qK9X/01q7H7D+zk5hDXP77UUD0+FSWk7h3vPYXbVhHCbMZOYubtzGGtf3h23D733qioLfVW5d2hUdxsist91daX49fH/woYNG8J1RutaQGRFZA2JbFuN1dFjk7aOHtz01G4cOBV6dpk/KAEAihyhzz5Vjnegr+7TWP7Tg3A6jsSNb9hGU4ufQMtyShztvu5BXPAHsa97EJt2ecI/d/aPoLN/JPzag6+ewNnxs/Avb53Gpl0eXPCHMnEjv/giA+jsH8HKbe1o6+jJyN+LiIgoVmTN8wUlXPAH4QtKONTnxbvTP4rdncYehKu1HyXSnbWPtKiNu8y2mbPjZ+GBV08k3Efse8WO6dhWiTLHjmsBkRq9thU7fvEFJcgI3b8rP58Y8sPnmgi/BI5vMoATaFnOTOQtEOpcPZM/iBc7zhjbP2NtiYgoS+jWPEHAc3v7k94PI91Jj1rbMdtmeirmG9pHojEe2ypR5thxLSBSo9e2zN77q+2DUotf4cxyZiNvAcBbPAlSQE68IhhrS0RE2SNRzes8M2rLflj7SItW2zHTZryuckP7MDLGY1slygw7rgVEavTalmzsFl4T22fq8RNoWc5s5C0AuEfPwO00FvPBWFsiIsoWiWpe3aRiW/bD2kdatNqOmTbj9p01tA8jYzy2VaLMsONaQKRGr21ZufeP3QelFifQspyZyFsAEABUD+zDzfMmGVqfsbZERJQtdGueLOPWBRVJ74eR7qRHre2YbTPV/fsQ2/zU9pFojMe2SpQ5dlwLiNTotS2z9/5q+6DU4gRaltOLvC1xOVBXUYK6ipLwaw/dMAPl57vw5Q9Pxd2L61HicgBA1EBOAFA3ZTx+0sJYWyIiyh6RNa94LNbd5RQxp9KNS0/+D66tM/abVa39MNKdElEbd5ltM+Xnj+HBG2Yk3Efse9VVlKBuynjL70tE9rHjWkCkRq9txY5fXE4RAkL3766xn2eUFcE1ehZFIlDicqCY45u04jPQcoASR/voywexdU8n9h0/CwGhwICugRHMnFyCJz67AMvmVcPj8eDFse0WzCzHy+8U49jACCADrrHoWyXmlh2MiIgyLTLKfXbVBCysnwLIMkRBwGVTJ6B1aQMa3Oex7nddcds98tIBHBsYgQBg5uQS3HNjYzi+nVHuZFWithPZZqvKQl8r7h0axeyqCfibxtDXb66rK8UdH11g6r2U/fYOApBlJPkoHKKCF1tfWpc2YNm8atXlDW71fQyO+DDiC2Lf8bNYua0dsyqiaw2RFXp1JlEN8ng8WLduMzZs2ID6+oufOGvr6MFNT+2Oa++Rr6v1BzKHn0DLEY++fBCbdnlwwR8EgPCgSpKBzv4RrNoWHVu7u3MYK7e1o7N/BLIcHX3LmFsiIsoGalHum3Z54qLdd3cOq26n1DitWkhkt9g229k/gs7+kXB7ffDVEzg7flbS++VYjSg5Wn3q0ZcPqi7XqzMKGaw1lJ0S1RDWGPtwAi1HbN3Tqfu6jOjY2mf39uuvz5hbIiLKMCNx7bIMPBdT07S2i62FRHZL1GZlAD2TP2jLfjlWI7JOq09t/X2n6nKjdQZgraHsk6iGsMbYhxNoOUL55JmeyNjaY2dGTa1PRESUblpR7rE6Y2qa3nasbZRKRtqst9hYkJOR/bI9E1mj1acu+NTvqczUGYB9k7JLohrCGmMfTqDliHFFjoTrRMbWzppUbGp9IiKidDMa114XU9P0tmNto1Qy0mbdo2ds2y/bM5E1Wn1qnEv9nspMnQHYNym7JKohrDH2yakJtNdffx1//dd/jenTp0MQBLz44ouZPqS0uXNhne7rAqJja29dUKG/PmNuiYgow4zEtQtCfE1rXdqgvi5Y2yi1ErVZAUD1wD5b9suxGpF1Wn3qzmvqVJcbrTMAaw1ln0Q1hDXGPjk1gXb+/HnMnz8fTz31VKYPJe2+vWIO7l5cD5cz+pQJAOoqSrBqcT2e3nEUn/j5IRyc+UkAwJaWJtRVlEAULkbfOkUB41wOFIkint5xlA8OzEFKwkrj917BTU/t5jkkoqyndd1Si3K/e0k95teWwykKEIXQMzp+sOMkTlRcFd7fsnnVUTVOFEK1cEsLI9wptZbNq8aq6+vDn2JxOUVMLS0Ot9+HbpiB8vNdCfaivt/YvrClme2ZyCq1PrVqcT32eN5HkSiipMiBYqcY7mvX1pXGba/UmUguh4hVi+vZN8l2ydzj6dUQJX2zSBQxzuWAK6Ldsx2b58z0AZixYsUKrFixItOHkTELZpbDt0uKXigAKy6fhk27Ih4A6K7Eg6+ewOaWauxctzS8WEnfCIx9919J39jc3MQI2xyhnEMFzyERZbtE1y21uPZHXz6IfcfPhn/2BmR4J38Q//LWafxwLLI9Ucw7USq0dfREjbl8AQl950bD7dnj8eBFi/tmmyayV2Sfiq1FvmDoEzitS0KTYR5P/MPUlW2jt5Ow+XUPFswsZ38l29hxj6dWQ2L3i5h2T+bl1CfQzBodHcXQ0FDUn1ymlZ7xC5WETrV0GKZv5D6eQ6Lskm91JhWsXLdUk6cFAS92mH+2FJGdWIcpE1hrkme177LPUzqkqp2x/dovryfQHnnkEUycODH8p7a2NtOHlBSt9IwRjYTO2FQNpm/kPp5DouySb3UmFaxct7SSp70B2ZZjIrKKdZgygbUmeVb7Lvs8pUOq2hnbr/3yegLtnnvuweDgYPjP8ePHM31ISdFKzyjRSOiMTdVg+kbu4zkkyi75VmdSwcp1Syt52u1MkDhAlGKsw5QJrDXJs9p32ecpHVLVzth+7ZfXE2jFxcUoKyuL+pPLtNIz7rhWJU0G8akaTN/IfTyHRNkl3+pMKli5bqkmT8syPjVvkr0HR2QS6zBlAmtN8qz2XfZ5SodUtTO2X/vl9QRavtFK1/j28jnh5W6ngJILp/HQDTPiHgzIhKfcx3NIRLnGynVLSZ4uGUs6dDsFVA3sw5c+PDVdh02kinWYKDdZ7bvs85QOqWpnbL/2y6kUznPnzuHo0YsPwnvvvfewd+9eTJ48GTNnzszgkaWPVrrGxh1Hcbj3HCpKnOgbKcb3f3sS6187BX9QAgRg1uQS3HNjY3h7ZZs1z+/F7KoJaF3awCSZHBB5rmdXTWCCShaJOzdLG9DgzvRREWUHrXRBtX6j1Kg9nvchy8D8mon4m8YJ+PXT/wrgc6r7V9sPANV9EyUrmbRMpa0eODUMhyggKMlonFbKNktkk7aOHjzy0gEcGxiBAGBmzD2QkT6lVlN+vfpaS+9JFEtr7JNsEnOq9kvRcmoC7Q9/+AOWLl0a/nnt2rUAgDvuuANbt27N0FFlVmw07Ql/EHBNBCQZoSzO0P919o9g5bZ2bGlpAoCkY3Ip/eyIN6bU0Do3D35sRgaPiii7afWbVdfXY9MuT9Tyt7sHccn4WYb3ExXZDl4vKTvEtlWMZWWwzRLZI7aPyQjdA63a1o7NLcb60u7OYTzw6sHwz4n6oh3vSYUjVfdzvE9Mn5z6CueSJUsgy3Lcn0KdPAPUo2l119/pYZxtjuJ5y15a5+a5vf0ZOBqi3KDVb7b+vjN+OYCeyR80vB81vF5Sppkds7HNEpmj1cdkGO9Lz6qM3fT6oh3vSYUjVfdzvE9Mn5z6BBrF04qm1XKkdxiyrP0aZS/GEGcvrXPTeWYUc9N8LES5QqvfXPAFVZd7i9UDBMzUQV4vKZPMjtkAtlkiM/T6mNG+dOzMqKnt7XhPKhypup/jfWL65NQn0CieVjStlsuqShlnm6N43rKX1rmpm1Sc5iMhyh1a/WbcWHBALPfoGVP7UcPrJWWS2TEbwDZLZIZeHzPal2ZpjN20trfjPalwpOp+jveJ6cMJtBynFk2rZ/WSesbZ5iiet+yldW5uXVCRmQMiygFa/ebOa+rilwOoHthneD/KNrH75vWSMinRmI1tlig5ShhHLAHG+9KtCypMjbfteE8qHKm6n+N9YvrwK5w5REl46RoYgYyLyZqbm5uwcacHHScGEZRCz4VTG6HVVZRAxsU42407PTjSO4zLqkqxmmmOWY/nLXtpnZt693m8aGB7rdQconwRmzzoD0oQBSH8SAEBwKyKEqy4fBr2eN6HQxAQhAxZBkqKHLhp7kQcOtyluu/I/nfw1BAcggDf2P79QQmCEEpE++6NjbxeUkbFtVVRQECS0TitDKuX1IefmcQaT2TNsnnV2NLSFL5fAsxd/8+On4Vn9/ajSBQBAfAHJEAI3XPJ0E7bTOY9qbAodSC2vWg8YSks8l5hZnkR/DHBSmpjoYAk4+kdR8P3/2r74n2HeZxAyxFxyU2ITtZceGkF9h0/G3pB49ebnf0jUWkc7Ci5h+cte6mdG48n8YM7mZpD+U4reRARw0UltSwyfVMx4g/i+X0DmimcwMWBYXSdDO1floFjY794Isq0RHWc132i5FgdK+/uHMa7Mz4G9HmjX5AvpmrK0Yuj0jZ3rlua1HFTYensH4n6bzNJr4f6gsD0j2J35zDqIz5gpjYWir2v4H1H8vgVzhyhl9y0cacHW/d0GtoP0ziIsgtTcyjfmU0eVKOXwmnkfdiniIhIj1r6ZiStX8IwbZPMMjv2Vx3fCAKeU2mzifbN+47kcQItRyRKeLngV08t01qfiLIDU3Mo31lJHlSjlcJp9H3Yp4iISItW+qYRrC9khtmxv9b6nSptNtG+ed+RPE6g5YhECS/jitRTy7TWJ6LswNQcyndWkgfVaKVwGn0f9ikiItKilb5pBOsLmWF27K+1fp1Km020b953JI8TaDmidWlDXDqTYvWSety5sM7QfpjGQZRdmJpD+c5MWrTWanopnEbeh32KiIj03LqgAuFkGxV69Yn1hcwwO/ZXHd/IcqjNmtw37zuSxwm0HLFsXjU2tzShrqIEogCIAlA3ZTx+0tKEj8+rxrdXzMHdi+vhdgqALIcv8gIAl1OEyylifm05tjQ3MRGGKIsoqTnza8tR4nKwn1LeiWzjxU4RJS4HnKIAl1OEgLF6VlGCn7Q0YXNL9HpK7XrohhkoP6+ewqn2Pq6x7YtZ+4iIyIDr6kpx6cnfYk6lGyUuB+oqSlBXUXJxbNbShC2x92IVJdjSwvpC5pgd+8euP6fSjUtP/g+urYv/1FiiffO+I3lM4cwhsakyj758EGue34sL/iDGFTlw58I6/L+/+wDWrVuHm1bfj/88cI7xtFnMaIQwo4azj93nhOmqlO8i23hk/2msmYiF9VOwx/M+1jy/V7U/tXX04PFXOnCk4Xa0vtiJtcvH6/cXWYZDEHDZ1AkJ902USqzfRKlntZ/Fbvc3jRNQfv4YNtxch/p67U/jsA9TsuLa7JL6uAmsyHWqykJf1ewdGg231V//TvuXilr3FbHv+8RnF7A9W8BPoOWoR18+iE27POHwgAv+IDbt8uBf3jqNs+Nn4YFXT2Bf9yAu+IPheNq2jp4MHzUplAjhROdod+ewofUofYyeOyKKp9Z/Nu3yaPYnZf1DfV5IYhEO9Xk1+5vZfROlEus3UepZHZOpbffgqydwdvysNB05FSojbTZ2nc7+EXT2jyTVVnn/Yh9OoOWorXs6VZe/2HEGPRXz45Yznja7GI0QVovU5rnMLMY/E1mnGsUew2rcutl9E6US6zdR6lkdk6luB6Bn8gftOjQiVUbabKLxjJW2yvsX+/ArnDlK+eRZLG9AhugqV32N8bTZw2iEsFakNs9l5jD+mcg6rf4Ty0rcutl9E6US6zdR6lkdk2lt5y2elPQxEekx0maNjGfMtlXev9iHn0DLUeOKHKrL3U4Bbt9Z1dcYT5s9jEYIa0Vq81xmDuOfiazT6j+xrMStm903USqxfhOlntUxmdZ27tEzSR8TkR4jbdbIeMZsW+X9i304gZaj7lxYp7r8U/Mmobp/X1zUMuNps4vRCOFbF1QwajjLMP6ZyDrVKPYYVuPWze6bKJVYv4lSz+qYTHU7ANUD++w9QKIYRtpsovGMlbbK+xf7cAItR317xRzcvbgeJa7QJ9FKXA60LqnHlz48FeXnj+HBG2YwnjaLGY0Qvq6ulFHDWYbxz0TWqfWfuxfXJ4xbn1Pphij5MafSrdnfVPe9RHvfRKnE+k2UelbHZGrbPXTDDJSf1042JLKDkTYbu05dRQnqpoxPqq3y/sU+fAZaDlAiZ/98YhCSfHH5xHFFkGVgfs3EcGSzxxPxIEBZhiwDgyM+3PfiO1i5rR0yQrPWlaXFKHE5wnG4jFZPPbWY7V+vvjbhdlpRxJQ5PCdE5rV19OCRlw7g2MAIZDlUi/YdP4vDPcO4c2Gd5vVw2bxqNLjPY926ddiwcgPq6xNHvcty6H8W1Jbj28vnpPhvRqRjbCwGWYYyhHv05YP46e734AtKAACnKEAUBDROK+V4jMgkpb9s3HEUh3uG8fSOo5AjluttF7mOx+PB1vGz0PpiJ7rOHsHsqglYWD8FezzvR43dI7dRG9uz/1IisW2vraMHSzbsQNfACGQAsyaX4J4bG/Hr1dfGjXEGR3z4wY5BiDM/id2dwzjqTdwGY/cxtbTYVF+haJY+gdbV1YXR0fiHo0qShK4uztzbKTJyNnLyDAAGL/hVY2jPjp+FB149ERV9e3p4NDxwkwGcHh6NisNljG1qMTqYiAqZcg3s7A9NngEI16QL/iA27fLg0ZcPWt6vVtQ7r7OUKbs7h1Xr/qpt7di0yxOePAOAgCTDF5TYZokssGuMvbtzGO/O+BgO9XnD+9m0y6O5X47tyQ6R4yNJDv0CsLN/BCu3tePRlw+qjnG8ARkj7ko8+OqJhG2Q4yT7WfoEWl1dHRobG/Gb3/wG9fUXvzfb19eHSy65BMGgekJkrunr68PQ0FBGj+HxVzoNrSfLwBOvdOCeq5zoqZhv+n2U7Rvc501va4eysjJUVlZm5L3TQS86mLP+RJTvEkWyA8Avft+Jb68w92mxhFHvvM5Shjy7tz9umSwDbfv1b1TYZonMsWuMrdZn9fbLsX16ZcN9eSro3ev//Hfv6m4rqy2LuadPNJeQ6TkAu6RzLsHyVzgbGxvx4Q9/GL/61a/w0Y9+NLxcltVOZe7p6+vD1776FYz6/Bk9jiMNtwNikaF1D/cO48kn/w3ehtstvdfh3mGsW7fO0rbJKnYV4Z9//FTeTqIxOpiICpmRSPYRn/lfvhnZL6+zlAnHzsR/UwMAjAyT2WaJjLNrjK3VZ7X2y7F9+mTLfXkq6N3re/0SEqYjqYi8pzcyl5DJOQC7pHMuwdIEmiAI2LhxI5555hl84hOfwD/+4z/ia1/7Wvi1fDA0NIRRnx/NM3tQ5fZl7Di+h168ixpD614i9uKbs7tMbaO2fbr1el3Y3lWNoaGhvJ1Am101Afu6B+OWMzqYiAqB1jUwkhKKY/d+eZ2lTJg1qRiH+rxxywUh8SQa2yyRcXaNsbX6rNZ+ObZPn2y5L08Fvfv2YsGPUbhM7zPynt7IvECm5gDsku65BEsTaMqnzL7xjW9gzpw5+PznP4+3334b999/v60Hlw2q3D7Ulhj7jUQqfDHwGu7zNUNO8Lg6ARK+4HoNtc5Rw9tobU/2a13agFXb26MGzYwOJqJC0bq0ASu3teuuc+fCOkv7jb22RuJ1ljLl1gUVePB/TsTV/WVzq/GKzvNm2GaJzLFrjH3rggo88N/dup/4idwvx/bpl+n78lQI3be3QEZsu5Pwaefv8GxgseY9vQAZoXgaMWJZ9D19onkBzgGYl3QK54oVK7Bnzx588pOfxFtvvWXHMVGERc79WI/t2ORfgW55iuZ6MgTc62uB4FM6kgDlm9FF8EOCA8Fwx5FRCi+8KIIfTrjhwy3OPVjk3J/yv0+hUqKDN+704OCpITgEAb6ghDXP70VQknWTt5TklAOnhuEQhYTrExFlm2XzqnH34nr89HfvwReQVNfZuNODjTtDSdJlbie8fgm+oIRxRQ7cNHcigNBDnr/4nztwbGAEAoCZk0uw6vp67Hm3H0d6hzG1NJTCeXp4FJdVlWL1knpGtJMlyabrXVdXis3NTbjvxT+jbzj0iYkiUcQlU8Zr9oVxLgfuvKaObZbIhGXzqrHq+nr8Yk8nRvxBCELoLsjIGDvSdXWluPTkbzHauAKnhkNfFayY4MJ4lzNcUxZeWoGndxzFmuf3YnbVhKj6w5pDRt3nvQ1vSPMgQ4AAGTPQh1OYjCCiP4n/bGAxJuEcSoRR9MtlqBBCz4Drl8tQJ/biQ6IHLweaMIDQpx4dkDAR5/B93+dQF+hFi3NH1FzCCbkCgAARQUhj8wLThYG0/t3zgaUJtMWLF8Pluvhxwrlz5+Ktt97Cpz71qbx5Blq26ZaNfRwxfvYa8Md89FMAMIyS8M9eFOPZwGLMFY9zEi2FlMId+SmMgBR65o+SgrK5uSmqwO/uHMYDr0Yk0409IkhrfSKibNTW0YNNuzyG1x/yBsL/fcEfxPP7BjCx+iN44NUT4eUyQklVm3d5sLmF10Kyj5JaprBac/d2nQ1PngGALyhh0y4P7l5cj8PrV8S9zwVfEJtf92DBzHK2ZyKDYuuLLIfqwwW//hhby4mhi8/Z6hv24X3Bh83NTQAQd114+8Qgx+Jkyn3e2/C6dEX4ZxkCujFVZU0BMoABlOGMLGO9a1vUffobgbm41xf93PMgHBhA6BeOB6Va3OdrxnpsBxA9lxCMmAI6IU8Jr8d5AGMsTaDt2LEjbtnkyZOxa9euqGU//OEPsWrVKpSXl1s6OArZFlhq6/7UJtlkiNgeWMKOk2J6iXFqyT16iUBM+iHKX/mWNmU0UVrPYOks1eUy8iNByqxsSa/Ot7YKqLdXo0ll3d3d4f//+e/U1/35797Fp2cXJfU+uSRb2irlJyMpz0bHzD0V8zW3VXtWAMfiZNYb0jzT28gQ4u7TjcwPKPf3avf+autxHsCYpL/CqecHP/gBPvOZz3ACLUmdUlVevU8hS5QYF5vckygRiEk/RPknH9OmzCRKa9MeAOZDgpRZ2ZBenY9tFdBur2ba2ZNPPgnvZV9QfZ6S1y9h3bp1trxPLsiGtkr5y0gaM2BszOx1lWtuq/UlK47FyYxEk1laYu/Tjd63d0pVht6T8wDGpXQCjV/ntEed2IuDUm1a3odSK1FiXGxyT6JEICb9EOWffEybspoOHUkQZM1BYK4nSJmVLenV+dhWAe32aradfQHqCWrFgl83NT2f2nO2tFXKX0bSmAFjY2a37yxG3PHt9LKqUkCWmbpJSQs9+N/8JFrsfbrR+YE6sRcyhITrch7AuJROoJE9Wpw7TKdqRlNCBUK0EjuanTuTOUwyQC8xTi25Ry3FS299Isof+ZQ2ZSUdOpqMRWIHXpfmATH7ECAzQSrD8qmtAurt1UpS2ad9v8MzgSWI/vSkjE87f4dal3pqOhPRiMxJlMYMGB8zV/fvw3szPobIXSnbygBTNylpobHM5YitC3qfshcgx92ntzh34F5fM2LHRNHbXby/T5TEyXkA46yOZCmNFjn3Y71rOxrFLrjghxujcCKAIvgR6nAX/wiQEHrS/MWf5wrH0Ch2YRxG0Sh2Yb1rGx4e29/FZXxwYDooaZzza8vhcooocTlQ7BQxv7YcW5qb4pJ7lBQvo+sTEWWjyDrmRGCsVsmaf0pxPlzj3BjFbc6dWO9+Bg+7tqNG6IMACQIk1Ah9cQ/WJUpWZHtNZpy00tWG25w74cYoItvySlebre9DVMjUxtZOUUCJywGXyTFz+fljePCGGZhfW44SlyNq28j3iX2NyKj17mdwvfhOeByk3KtfHNvIcCIAJBjnLHLujxsTVWAINUJfXD2JrTU1Qp/qemQMP4GWI5TG/0ZgLjb6b8RJeTIAYIbQj9ailwAgvFyAgOljyyO38aIIB6SasRnoUGyuCMnyd7HJmmVjRdjq+m0dPXjkpQOh34IBmDW5BPfc2MgHmBJRVlPqmBFvBOZiW2ApOqUq1ImnAQB3eVeP/dyLRUIH3pJmo3ssPUr2hWradGEA9cIpvCXNhhcuuOHDLc494QkLIqMi26vSHr/v+xzqxF60OHeEx1cX2+nF5ZHmisfRLp4Ot+W54nHN9yEia5SxsjJGPtY/ghFfEAKAwREftD6c1tbRg407juJw7znMLC+Cf/wsXFdXijs+uiD82prn92J21QS0Lm2Iep+NO47iK8/+CQ5RQFCS0TitNLwOkZ717mdMrf9GYG7UGEipNbH1Q6lJ78sTMSSXYKP/xqi61ezcGVWz7i56mfXHAk6g5RC1uNoT8hTc62tB7Mc+Q5G0LbhV2olnYlI65PD/CwhCjIq5ZSfKbrGR9wDQ2T+CldvasaWFMdpElPtia91BqTbq2R0HpVocxMWflV8CyRBwQp6CE/KU8GteFI99hQ6cRCNL1Nrjfb5m3CrtihpfqY2ltLbleIvIfmpjZBmhcfKqbe3YHDNOjl3/UF8QmP5R7O4cxlFv9Gv7ugexans7Njc3AUD0+wTj1+F4nOxitI7Erhc5Fgpt0xL1oRnWI+tS+hXORYsWYdy4cal8i4KiHVer/gkyGQJeCCw0tG8lvpaym15U98adnjQeCRFRahiJZjfHeC0kiqXWHmWIqm0qdiyltS3HW0T20xsjy4gfJ6uuLwh4bm+/6muyHNqH7vvIHI+TvYzWkURjJ7VvnLEeWWP5E2iSJOHo0aM4ffo0JEmKeu36668HALz00kvJHR1FsRIv61VJf7Jz/5ReelHdjNEmonyQilpkphYSRdJqj1ptKnJ9rW053iKyn94YGYgfJ2ut33lmFILg19yHXliB2vsQJcNoHbFaV1iPzLM0gfbmm2/i1ltvxbFjxyDHXEUEQUAwGLTl4Cia0bjaSG744EWx4f1TdtOL6maMNhHlAyu1LhE3fLbujwqHVnvUGl9FjqW0tuV4i8h+emNkIH6crLV+3aRiFBcXq752WVUpIMum3ocoGUbriNWxE+uReZa+wrlq1SpcddVVeOeddzAwMIAzZ86E/wwMDNh9jDSmxbkDgKTyiqy6XICMW5x7IGg+OjNyXcbX5oLWpQ2akQ+M0SaifNDi3DGWThUpto7p1bX4dW9x7kn+wKggqbVHAdLY+Cp+eeRYSmtbjreI7Kc3RhYQP05uXdoAIXYDWcatCypUXxOE0D5Ut4tZh8guRuuI+tgpcpv4+QLWI2ssTaAdOXIEP/jBD9DY2Ijy8nJMnDgx6g+lhlpcbY3Qh4dd21SXr3dtw0pXG9a7toVfUyJxldhcJwKMr80hy+ZVY3NLE+oqSiAKgCgAdVPG4yctjNEmovwQG7feKHbhNufOqJ+vF9+BG6NQapny/zVCX9RrboziNudOBgiQZWrtcb1r+9j4Kn555FhKa1uOt4jsFzlGVua3BAB1FSXYojJOXjavGpubmzC/thwlLgfmVLpx6cn/wbV1pXGvza8tx5bm0D4iXyt2iihxOeByilHrENnFaB2JXa9G6EON0BexTWi+gPUoeZa+wvmXf/mXOHr0KBoaGuw+HoqxxbcMLwQWwgsXBMgQIMOFAJwIol7siYpSnyB4USwHUCEMQYYQiq0NhGJrnx33mKG4dcpOkTHbs6smYMXl07DH8z4O957DRLfTwGcMKV3UItFjXztwapix51RwzNQgtWj2dqkBAYg4LM3AgbEUziIE8Bnnbqx0tYX3/5b0AdY4suSNwFxs9N+Ik/JkAMB0YQCtRaHn+crhUdjFj54o7Wuj/0YclGpwn68Z0/2hbZTXlP/fFliK96Tq8IOe2TaJ7KWMsXqHRvHBmolxY6u2jh488tIBHOsfiRs3uxwijvR7IU9bguU/PYi500+hdWkDfr36WtX3UvarjPcaq0vRuqSek2eUEsqYSBnnRN7jx/7SJra2RG0j9qJJPAoZAutREixNoH31q1/FN7/5TfT09OCKK65AUVFR1Osf/OAHbTm4QrfFtywqIl0e+/ClFw4A2lHq8bG1xuLWKTvFxmzv6x6MevYCY7Ozh5lIdMaeUyExGsNuZNtIfhThmcASHJem4HXpCkv7JwLU29kJeQru9bUgMu08sm0BUN3mPl8L1mNb+IbHatsnImPUxsqRY6vY12P5gmNfbROLIEmJx2aJ3o/IblZqido2kc9JYz2yxtIE2i233AIA+MIXvhBeJggCZFlmiICN1CLSY2lFqRtZR4muZYfJbnpx2QolNptFO7P0ItGLi7XToXj+KN/pxbAnqkGJotkBAW9I8yzvnwjQa2fxDztS2pas8cQlGUK47SXT9olSra+vD0NDQ5k+jKQ9/kpn3DJZBp54pQMN7vOqrycSub3Z98sXZWVlqKyszPRhEKyNoxKPn1iPrLA0gfbee+/ZfRykQisi3cp6RuLWKTsliuVWMDY786xEoit4/iifGY1ht7qO1kQGaxwZZbatdEpVmu0ucn/JtH2iVOrr68PXvvoVjPr0xye54EjD7YBYFLf8cO8w1q1bp/l6Isr2Zt8vXxS7ivDPP36Kk2hZwEotMVpnWI/MsTSBNmvWrMQrUdK0ItKtrGckbp2yU6JYbgVjszPPSiS6gueP8pnRGHYz20YSIKtOZrDGkVFG2lns+jIEzW2UtpdM2ydKpaGhIYz6/Gie2YMqty/Th5OU76EX76ImbvklYi++ObtL8/VElO3Nvl8+6PW6sL2rGkNDQ5xAywJWaonRusZ6ZI7hCbTf/OY3WLFiBYqKivCb3/xGd91PfvKTSR8YAbc49+CZwBKofX1AoUSpPxtYDFkjVFVrHUbX5obWpQ1Ytb0dsk5SAGOzs4PquRqLRK+qrtY8jzx/lO9anDtwn6/ZUg1S2zaajEViB96Q5rHGkWUtzh2419eM+IB6eeyPettS20aAHH49mbZPlA5Vbh9qS0YzfRhJ+WLgNdV+9gXXa6h1jo693qL7qdFYkdubfT8iu1mpJerjJxmRcwusR+YZnkC7+eab0dPTg6lTp+Lmm2/WXI/PQLPPSlcbAESlcF688MsoGnsK+c7gFZiEcziDCQBCAzdprKM4IaEM5/GrwHVwIgh/RIdxIoj9Ui0Wgd95zpTYdE21NEYlLnvjTg+O9A7jsqpSLKyvwB5Pf/jn1Uz+yQqx52rmxCL49v4Xrq37KurrL7524NQQnGMpnHOmlfH8UV5TEqBESJAQ+rqlGz7c4twDALjLuxoeqRoOSAhCxFQh9EnNfrksnKa53rUd2wNL4JGmQYKAwFiYjgAZImScRjkWiR34vTQH/rGhzSQY+/o7ERBKIXsY2/G47yb0owzAxZTXueJxbA8sCSfINjt3hp8X8zC2Y5N/BU7IFQAEOBEAgKh0s/XYrrk9ESVPrZ99SPRgo//GsSAQhfpvo4sQgAwBTgQhQUS9eCo8qXCXd3VcejT7NaWaWnK5MhbSa3Ox2y0SO/CWNBteuOCGD5cKPXhXrg7/fItzD9utSYYn0CRJUv1vSq2VrjasdLWppEMJ8I9NkkWmbgLRpSEABwbGBoKxlPQy5X0ovfQSfBrc0esum1fNB8zniMhz5fF4sO53XaqvERUCrQRNL4rxTGAxIj+5ozyFRy1Jer1rO7a4N8btV4aA4Nh6BxH9NYUBlOFeXzMeZroUmdCPieH/9qMIzwYWx7W/SMrNtNIm/Qg9Fyk23YxtkCi1IvuZXnpzLAESHnQ9pzoRoZd6yH5NqaLZ9nRqkdZ2kWMjL4qxX54V9fOzgcWYKx5nWzZB6/sQlGWMpGhYIxhK+yT7qSU2KmmMRET5QL92GRuCKAlRxvcb/R6x2xJp0Us5S+W2RGQvM/dNWv2UfZoyxWrbszJfwDZtnqUQAQA4f/48du3aha6uLvh80Q+e/NrXvpb0gVG0VKZjGE37JHtpJTaG0hj5KSUiyn121a7Y/ZjZL9OlyKhUpMWy/RGln5VUXaP7YJ+mVLPa9qy2TbZpcyxNoP3pT3/CjTfeiJGREZw/fx6TJ0/G+++/j5KSEkydOpUTaClgNh3KDDdyO3knV2klNjKNkYjyhV21KzYhysx+mS5FRqUiLZbtjyj9rKTqGt0H+zSlmtW2Z3XMxTZtjqWvcH7jG9/AX//1X2NgYADjxo3Dm2++iWPHjqGpqQn/9E//ZPcxEkIpGgJS8ew5OfwgZ0qv1qUNEGLCgJjGSET5RL92SRA0HugcSS0hSn2/avtiuhQZp9auzKTFWt2WiOzV4twBaNae6Fqh1U/ZpylTrLY9Y2MjY+2ftFmaQNu7dy+++c1vwuFwwOFwYHR0FLW1tfjHf/xHfPe737X7GAlj6TKu7WgUuzAOo6gR+lAj9Kn+dwUGIUCCAAkVGEKN0AcX/HBjFCKCYx1LhhujuM25kwECGaIkNs6vLUeJy4H5teXY0tzENEYiyhuRtUupQy740Sh24WHXdqx3bYt7LbKmNYpdWO+KDwGIrYmNYhduc+5EjdAXrn81Qh8eVtmWSItau1Jrf3ZvS0T2WuTcj4dd21Ej9CE0YRD6UwQ/rhffMdRP2acpU6y2Pa2xkd7PbNPmWfoKZ1FREYSxj85UVVWhq6sLjY2NmDhxIrq6uhJsTWbERtE2iUchQ4BHqoYDEvxw4H25DEGIqBd70CQexc7gFeHJZbfgw91FL2tG3L4QuBbtUkM4lpnSi6mM+aWtowcbdxzF4d5zmF01AX/TOCHTh0SUcbFJZUr9+b7vc1F1rVOqQqUwCBkC+uWyqIh2tTh3tQS0lWjTfB/WOUoktp0p7c+o2Da5xbcM3x/5LLxwwQ0fbnHu4S8tiVJI6cMeKTS2DsAJATKcCAIA6sUeLHP+yXC/ZtImpdsW3zK8EFhoqm5ojZESrf+eVB0OHmA7N87SBNqVV16JP/zhD5g9ezaWLl2K+++/H++//z62bduGK664wu5jLFiqUbQR32v2j/1/cOw0xr4OACfkKbjP14L12KYZ7Rwby0xE5rV19GDltvbwz/u6B/F29yAuGT9LZyuiwpKorp2Qp0S9dp+vGbdKu/BMRLKUkZrFOkdm2d1mtviWRbVbL4rxzFjSGSfRiOwX24cvEuAf+9IVawFlMyt1w2zt4vgoeZa+wvmDH/wA06ZNAwB8//vfR0VFBe6++26cPn0aP/nJT2w9wEJmJYpWjQwhKp6WscxE9tu442jcMhlAz+QPpv9giLKU2bomQ8QLgYWqy/VqFuscmWV3m1Frt4CgsZyIkmW0vrAWULayUjfM1i6Oj5Jn6RNoV111Vfi/Kysr8dJLL9l2QHSRnZGykftiLDOR/Q73nlNd7i2elOYjIcpeVuqMFy7T+2KdI7PsbjNa7VZrORElx0xfZS2gbGSlbpitXRwfJc/SJ9AoPeyMlI3cl9Z+GWFLZN3sKvXnnblHz6T5SIiyl5U644bP9L5Y58gsu9uMVrvVWk5EyTHTV1kLKBtZqRtmaxfHR8mz9Am0/v5+3H///dixYwdOnz4NSYqOSx0YGLDl4NRs3LgRGzZswKlTpzBv3jz86Ec/wqJFi1L2fpnU4tyB+3zNkKPmOWUAgs5W8a8LkKPiadX2ywhbynZ9fX0YGhrK9GFo+pvGCXi7ezAqHFoAUD2wD93dl2bqsCwpKytDZWVlpg+D8pDZuiZAwi3OPXg2sNhUzWKdI7PsbjO3OPeMPbsmsm3LuMW5J5nDJCIN6vUlHmsBZSsrdcNs7eL4KHmWJtCam5vh8XjwxS9+EVVVVeFEzlT75S9/ia9//evYuHEjrr32WmzZsgUrVqzA/v37MXPmzLQcQzotcu7HemzH9sCScKrGh0QP/ijV45A0A5JKgRAhYTy8OIdxkAGIkCFBxL2+FhT5AviMczfmiscxXRjASXkyAGCG0K+a1En2i01pbF3aoJrCubtzGAdnfhKf+PkhfKD6FFqXNgAAHnnpAI4NjEAAMGVCMUpcDvQOjeruKx/09fXha1/9CkZ9/sQrZ9Al42ehZ/IH4S2eBPfoGVQP7AMg4O7/7z14XWfg9p1Fdf8+lJ8/hrPjZ6GnYj68rvKo5dmg2FWEf/7xU5xEI9spdW2TfwVOyBWQw4PE0NSzCAkyxLGfZMgQ8Gxg8ViKmh/+sWGLDOBB3+ch+UQUIRBOoo5M54x8HwCYLqTul3uU+7TGXGpJrokSz94IzEW71AAHgpAgQoYQlaZmNjGNiPQpfcqJIBzwIwAHBMgIwAmMpXDKEOBEEEGI2BZYiv1SLdqlBnRKVagQQr+gPS1PhANSuKY0iUfD6yip0TuDV+CkPDmqfs0QBtBa9BL7MSWkd/2fKx5HBYbRj9KobXYGr8DcwHHV9hVZuzzSNIhj7TcyXTP2PW917sIfpXrLidOFztIE2u7du7F7927Mnz/f7uPR9fjjj+OLX/wivvSlLwEAfvSjH6GtrQ2bNm3CI488ktZjSRe1+OQtvmU4IKlPGEpwYBgluF58B69LVyDys4F+FKnMaiN8c0GppZbSuGp7OzY3N0VNfLV19OCBV08A7kogIGNf92DUdkDo5vH08GjCfeWLoaEhjPr8aJ7Zgyp3Nn/9pQvAG6H/HAf8YcYcPIHPh18dcVfivRkfwV9hN/6I6+OWfx2/xFU4mOZjjtbrdWF7VzWGhoY4gUYp0y2rty0JjoifQrVKhgBZ5TUlVS04tlwtSSryfUKp1EyaIm2RYy6tpLJEqbBqSYACJHzP9UvV15mARpSc2D7lR6jPrXdtj+vPgbFbX70EaOVXtbHrxP58kYAT8hTc62vGw+zHpEPv+g9AI0VWGb+0YD22aU6ixW6vV7MOSTOi+geZY2kCbc6cObhw4YLdx6LL5/Ohvb0d3/nOd6KWf/zjH8eePeofaxwdHcXo6MVJBitf/+r1FpneJtX+A9cmWEPAG9I8zddiyRDxM99HUOfzJH1sVqT63zhbvvr3+CudcctkGXjilQ40uM/rrmeE2r7SLdVf/aty+1BbMpp4xSzxfe91QPQ33CFDxKv4y7h1ZYh4RbwWn3LvS9PRkR3sqDOFxq6EaTVKktQi537dpCkOGikRrfajlwprpN2xXZIVrDXajPSpVNadi9iPSZ9eW5U1HmVxcT1Bt31ZrVlknqUJtI0bN+I73/kO7r//flx++eUoKoqeACkrK7Pl4CK9//77CAaDqKqKToioqqpCT0+P6jaPPPIIHnrooaTed3vXtKS2T4XRy4r0H4MGQJaFhOtEek+qwmNH8+9rsNn01b8jDbcDYvxk4eHeYaxbty7hekbE7ivd+NW/aFqJNlZSBSk72VFnCk2q27myfyZNUTKsXr8TtTu2S7KCtUabkT6Vrv7Ffkx69Npqogk0ve31XuM9h/0sTaCVl5djcHAQH/nIR6KWy7IMQRAQDAZtOTg1sc9bU95TzT333IO1a9eGfx4aGkJtrdpHb7U1zzyFKnfmJ18ifQF+jCaIQRcE2VBHVFwi9uKbs7uSPTRLer1FKZuozKav/n0PvXgXNXHLY//ttdYzIrPnkV/9i1Un9qp+3N8NH7woVl2fcosddabQaPULO/ev9z7sZ2SE1et3onbHdklWsNZoM9KnUl131N6TKJZeW5UhJGyjidLHec+RHpYm0G677Ta4XC48++yzaQsRmDJlChwOR9ynzU6fPh33qTRFcXExiovjG4wZVW5/1n1l7NO+36k+y+wiGYvEDrwuXa6yjjz2Jzp54wuu11DrzK6/p52y4at/Xwy8ppp6Evtvr7Ze6HuAAvQ+VlgI5zHXaCXdWEkVpOxkR50pNEaT0qyI7EdMmqJkWL1+J2p3bJdkBWuNNiN9ymwCtPY6etuwH5O+RG1Vb2wkQLaUPs57DvtZmkB755138Kc//Qkf+MAH7D4eTS6XC01NTXj11VfxqU99Krz81VdfxU033ZS248gGK11tAIAXAgvjPpZZhFDSppLy9Ljv5nCSh/LaXPF4VMoUkzfSQy3hS+3fXms9AFGJcpNxDuOEUfTLZTyPWUrvnLMfUqGK7BdHpOkIjqUUFiGAMoxgBG6UYBRDGAc/nBAgwwEJ1cIZXJBdGEApZABFEalqEkTUi6ei+pHRay6RGqvX70Ttju2SyF5G+pRWyq6SRHgxhbMcIqRwTYlcR9lmV/DyuBTpGqEfdxe9zH5MuhK11cg0TRlAYCwkaYaB9sV7jvSxNIF21VVX4fjx42mdQAOAtWvXoqWlBVdddRWuueYa/OQnP0FXVxdWrVqV1uPIBitdbeFJso3+G8filEOeD1yPXwWugx8OCAh1uiWOP6NdasALgWuxc6xIhJLNUv/pQbpILVVVK844Mnb4+77PoU7sZXHOQWrnXG85UaEYkksQHPuN6AyhH61FL2n2CeVa+L48ER8QT0TFvit18KBUg/t8zZjuHwjvi/2MkpHo+h1VowPR9Vt5faP/RtznawZ8wHQhum1G0hoLEFFiRq71sf3y2cBiAAKcCKBXLke92INLhYN4S5oNH5w4Kk1Dr1SOEbijJyMCx6P6apN4FO1SQ9x1gEiNXltNdswSmyQdeQ/JdmkfSxNoX/3qV7FmzRqsW7cOV1xxRVyIwAc/+EFbDi7WZz/7WfT39+Mf/uEfcOrUKVx++eV46aWXMGvWrJS8X7ZTi0r3j92MBMdmrGWEom8j42sjo5oZn55ZZuKMea6IKB+o1S69iHazse+J4t6J7KDXLpWbdKNtM9G+iMge6vdOofvYg1ItDqI2avlAxGv3+Zpxq7Qr6p7qoFQb9dwp9l3KBqwpqWVpAu2zn/0sAOALX/hCeJkgCGkJEWhtbUVra2vK9p9L7IpkZpRt5piNM+a5IqJcp1W7tCLarcS+J4p7J0qWXrtc5Nxvqp0n2hcR2SOZeycZIl4ILDS0HvsuZRJrSmpZmkB777337D4OssDO+FlG2WaGlThjnisiymVmY9itxr7zWkmppNcu9V5Xey3RvojIHsn2qdhnT6fqfYiSwZqSWpYisGbNmqX7h9LDzvhZRtlmhta/e53Yq/saEVGuShTDbnR9vetkovchSlaiGm2mbbLeE6VHsn3KDV9a3ocoGawpqWXpE2gA4PF48KMf/QgHDhyAIAhobGzEmjVrUF9fb+fxkY4W5w7c62uGsXlQ7dhlRtlmjtk4Y54rIsp1WrVLK6I90XXSzL6I7JKoXZpp54n2RUT2SHzvpH+/dItzD54NLI7qq7HbsO+mTq+3KPFKhBXYjUP4bFxNWS7txvGR4gweWWqku11YmkBra2vDJz/5SSxYsADXXnstZFnGnj17MG/ePPzXf/0XbrjhBruPk1Qscu7Hw9iOTf4VY3HKQBGCkCFAgByVwrnY8U5cVHO/XMYo2wwzGmfM2GEiyhextQvQj2hPdJ00sy8iuyRql2baeaJ9EZE94vtlKIVTAFAvnkIlBvGWNBteuFCEAMowEp/CKR6P6qsfEj3heyz23dTa3jUt04eQI0ZwyfjX0DP5g/AWT4J79AyqB/Zh1/kL2IWZmT64nGdpAu073/kOvvGNb+CHP/xh3PJvf/vbnECzmVq0+X6pFi8EFsILF9zw4VbnLqx0tenuo11qgAwBEwRvOHL5Pak6/KBBXuzTK/a83uf6ZTi56y7vas0o+y2+ZVHn/hbnHt1zT+mh1k/Zp4i0+4Za/4hdV6lVsdfJyHU9UjUckOBEEPViD/seJcXMtTy2Hb8RmIvPX/h7nJQnAwDGw4tRFEGGADd8WOx4x/C+iMic2JoQhKhaE9T6mrLt76U5cECCA0E4IGEIJagXe+ImxWQI4T9qP1NqNM88hSq3P9OHkZQ/YA5+jUU4gamYgdO4CW/gKhxMwbZdAN4IbTNuEU7MWAyYfL9c0estSuvkqqUJtAMHDuBXv/pV3PIvfOEL+NGPfpTsMVEEtRjae30tiPyosBfFeCawBABUJ1LU9sHI5czSihdWi8eOPDdbfMuiXk907ik9GBdNpM5M3zBTqwBEretXWYd9j8xK5loeuy0AnENJ+L9Zr4lSJ7b/makJWtsGx26T9WoP76nSq8rtR23JaKYPw7I3AnPxhO/z4Z/fRQ1+hM9ivctYjTG7bTLvR9osTaBVVlZi7969uOyyy6KW7927F1OnTrXlwChEPW5Z7bcbAl4ILFQdlBmJbGa0bXppxQurxWNHnhv1+Gztc0/pwbhoInVm+oaZWqX3W372vfTLl+fS/BQfiVsmQ8TPfB9Bnc9jett4Av4jcC1uDOy0doA5KF/aRj7Lh3Ok1/8S9WEjfVfZh2zgWIxeM7JdPrSLbJPM/YKVbXl/khqWJtC+/OUv46677sK7776LhQsXQhAE7N69G48++ii++c1v2n2MBc1M3KxWtLLRfTDaNn20/q0TnUOt143GalNqMC6aSJ2ZvmGmViX6mgz7Xnrly3NpOhuqVJ8t/p5UhceO6j83RmvbWKNyER47wmfQUPbIh/6bqP/p9WGjffc9pa4YXDfRNYMKTzL3C1a25f1JaliaQPve976H0tJSPPbYY7jnnnsAANOnT8eDDz6Ir33ta7YeYKGrE3ujPhqsRyta2eg+GG2bPlrnxA0fvIhPR1HOjdbrRmO1KTW0zif7FBU6M33DTK2SIeiuy76XXvnwXBoA+B568S5q4pZfIvbim7O7LG0bq1jwJ9xXPkn3s2nIvHzov4n6n14fNtp3LxF7IQOG1831fs6+a79k7hesbMv7k9SwNIEmCAK+8Y1v4Bvf+AaGh4cBAKWlpbYeGIWoRZsj/AFiIWrZLc49JvfByOVM0YqsV4vHjjw3tzj3jD1Dxdi5p/TQOp/sU1TozPQNs7Uqfl39/VPq5PpzaRRfDLym2l6/4HoNtU79v19o25YEn46U8Wnn71Dryv1/K8of+dB/1fquIlEf1ts2dh+AWu2Jr1NGrhlUeJK5X7CyLe9PUsPSBFokTpyl1iLnftwq7YpLXTwuTcEb0rzwQE2AhGcCi/FMYDFEyJBiisBkDKNEGEW/XMbI5QyKTAhyIgg/QiU3Nto+9nwry5XnnDGFM7sscu7HemwPx5pXCEOQIeD7vs+hLsBETipcSg37VeA6+MeGHOPhxeO+m3GfrxkA4EQQAFAv9uBW5y7sCl6OE3IFAGAShlEi+MK1K7JWRdZGJf/MAQl+OHCfrxnT/QNoLXqJfY8M0xpzxSZhR7bnyRjGN10vjtWBbdjkX4ET8pSxX3WGRmlKCifrNVFqxPZdpR5cJp7UvMeJTNydjHMYwjj44QzXExcCkCCiXjwV3scbgbmYLgzgpFwR7uOTMQwBAgYwAQAwXehP69+dckt0+wmNgR70fR5T/YMAgNPyxLgU2f1SLV4ILIQMAQKkqLYNAHd5V6smR8fen/Ce3x6GJ9CuvPJKCIKxaN4//vGPlg+Ior0RmKuSurgYsV/Al+EI/7eksp8BlGFAlvAwUzcyRi2hCwj93kq5WVQ7388GFmOueDxqEo0D8OyjxKIzkZPoothrGhBKJjwX8bN/rJ4dlGpxSKqJ+gTPACbijCzFJUbF7jd0uwNIY7UwdF2dgvt8LViPbex7ZEiiGhybhA2Exlf3+lrwcEQ7u1gDhLHPpkj4nuuXbIdEKaJWE4KA7uRZ5Fgt8lnCoRoU32fVx/ECBjAxaskJuZLjPoqjdR/oRyis4YQ8JWJZyEGpFvf6WhD5CUcZAgIQ8CExFFKR6J5DuT8h+xieQLv55pvD/+31erFx40bMnTsX11xzDQDgzTffREdHB1pbW20/yEKmnkpm4OmVqpi6kUl6CXN6yXJMS8ktTLwhu+RDApaxZMKLtK6BsYlmRvcrQ8iLNLRY+dA2slGi67d6EjYACOF1WAOI0s9sv0uU+qy2rZGkaCPvTYXJTPuJpvYBJgEvBBaiXTwd9wrbXuoZnkB74IEHwv/9pS99CV/72tfw/e9/P26d48eP23d0ZHtKBlM3MifRv71eshzPW+5g4g3ZJR8e3ms03SyR2EQzM/tlGhoZlej6rZd4razDGkCUfmb7nZXUQ7N9mH2eItndHrxwsd5kiKVnoP37v/87/vCHP8Qtb25uxlVXXYWf/exnSR8YhZhJ4TS6P8qMROdSL1mO5y13MPGG7FIIyWhGxSaamdlvPqShxWI6Wmokun5rJWFHrsMaQJR+Zvudkfur2G3N3pOxz1Mku+/p3fChTjzNepMBlibQxo0bh927d+Oyyy6LWr5792643W5bDoxC1FPJJIQ+zmnsmXSR2zF1I3PUz2WIXrIc01JyCxNvyC75k4yWKJnwIgEyZMhAghRE9dS06CQ0ZX9MQyOjEl2/1ZOwAUAOr8MaQJR+Zvud3phca1v9bbQTo4mAxG1OmxI3IEQtu8W5B3PF46w3GWBpAu3rX/867r77brS3t+Pqq68GEHoG2s9+9jPcf//9th5godNKzwAwlvQUevi8OJbWEfrv+BTOGuH9qJRHSr/Ic+mRpkGEFJfuA4BpKTmOiTdEF0UnE4bq1WQMAxGJZU4EIQDhayGAhP1HrZ99SPREJXjGphsTJZLo+q0E+ESmcFZgCGtdv2bqGVEGme13aunpAFQTn9W2iR3Hf0j04I9SPfs8aVJrPwE44EQQEkRMFc4CAE7L5XH3iEoKZ2Q6tFKPWG/Sz9IE2ne+8x1ceumlePLJJ/Hss88CABobG7F161Z85jOfsfUAKUQOhyoL2C/V4uVAEwZQCuBixtMc8URUdO0bgbnY6L8RJ+XJOCFX4DHfzdjovxH9cllcoYjcjuwXGZVdJ/biAddzukU98hxuCyzF932f4znLIrHnM7bfRb7GQkaFTOkP78sT8YGxGgUAG/03AnKoevnhQAXOYUguwfd9nwv3KQAXr3/+6Otfk3gU7VJDXD9bCSYUU3ISJZZFJmFH1ui6wMVaYCT1TNnWI1XDAQk+OMPjPCA00fxN14usH0QGxfa7NwJzcZd3dbiPBSGiXuyJqkMn5cmQIeCEXIHpwgDuc/0SwMXaEzvGU3uPbYGl+FXguvB7GP3ENRUepe0o9wlVY5Nm/XIZJghe7Xu7ANAunh4b85zGXPF41D5jt4m9F4kcM/F+MnmWJtAA4DOf+UzCybLnnnsOn/zkJzF+/Hirb1PwYiNvD0q1cd91VqKaI6NrAcRF5Q6gLPwp0MioXLXIW7KP2jk08u8dux3PWXbQO59A4jhpokKh1lfu9TUj/un/Avoj6pPaerHXv8g6yH5GmWC1tqttq/akwwGU4V5fMx5muyYyTauPhfpp/GMFQpNoU+Jqj16/1n8P1iSKZ+XezmytSTR3wPvJ5NmQjaVt5cqV6O3lQ+ySYTbyVomutbod2U8vWtvsdmb3QfbTO59WzzVRPlK/hhkddpgbnrCfUbolc703PkZjuyayQq+P6X9CLL72aPVr/fdg36V4Vu7tzNYazgGknuVPoBkhy3LilUiXlRjaTqnK0seHGXmbGlYjhq1EbFPq6Z1PrX7H80SFKN3tnv2M0slqbTe6jpV1iSjE7n6jtj87xvFUWKzUB7O1xurcARmX0k+gUfKsxNDWib2WtyP76UVoW9nO7DpkL73zafVcE+WjdLd79jNKp2Su92baKts1kXl29xu1/dkxjqfCYqU+mK01nANIPU6gZbkW5w4IkGKWan+yT4muDW1n/BOAjLxNHbVzaOTfW/3cm9sH2U/vfFo910T5SP0aJkGvhkWup1/Dol9jP6N0S+Z6n6i+X8R2TWSFXh8L1Rat/hdfe7T6tf57sO9SPCv3dmZrjdW5AzIupV/hpOQtcu7HrdKuqOjaD4uH0SHNDD10GaFC4ICEauEMZCCc2DgBFzCMceF9hSJxBWDsa2ZFCEAAwhG5fHhgapiN1o5MTpkuDECAHJecWiEMQYYQl/xFqacWQx2EiG2BpWhx7sCtzl1xUdOLnPt1kzuJ8pHate9Dogc7g1eEk89CZDggIxj1Oz0B4li0OyDAAQkBCBAAzBD6canQg7ek2VH9DADu8q5WTZ1inyO7qY3PjF7v1epIbApnBYaw1vVrtlkiAyL7nDJediIIB/wIwAHnWD1R7nkAYJN/BU7IFZAhQICMGUI/7i56GQAMjdnV+nHke7DvkprpwgBOypMBhNKWxwm+cCJmbLtR2nVsW/bDge/7Pgu/zwkREvxwAACKxtq5CBnBsUmzIgRwjXgQfZiomsLJtmoeJ9Cy3BuBuXgm4mGAXhTjDWke1rv0EzoiEzYU0ljnUvhRFJ51ZsdJLSOR9oD6eRQgRZ3vZJK/yB7Kv3OihEEvivFsYDEARPVjnjMqFJHXvthrV4gEQERQZdtgxBAlMFa/ZISui91yZfg1L4rxTGAxYpPTmNRJqaQ2PjNzvTc6LiAifXr3QH6EPmHzkOvZuP6m1/+M9k32YzJKbQw0gFKsLzKWpqmkvAbGxkbB8P87ItaJ/3KhH0V4Q7oc613b2FZtktIJtFmzZqGoqCiVb5H39JI3IjuB2cQNvX1R5hg530bbRD7q9WbP9eSn+IjKUvX0pv8IXKu6/Ge+j6DO50nB0VmXTf/GlF+SS+W8SD2sI/F+CuU6SemhVYtfCCxUXc62R5QaRpMN2f8ok8zev1m9t1cjQ2AfsJGlCbT/+7//gyRJ+Mu//Muo5f/7v/8Lh8OBq666CgDwzjvvJH+EBc5owkYy6RlM3sgeRs63nakruWZ717RMH0JYZ0OV4Xv/UbkIavf870lVeOzoTHsPjChLZcM1KhuOgfKDVlvywmVqfSJKDlPrKRekI03TyvuTeZYm0FavXo1vfetbcRNoJ06cwKOPPor//d//teXgKJSKEfk1lMjlRtYz+h6UHYycb6NtIh81zzyFKrc/8Ypp8D304l3UGFq3WPBjVOWm6hKxF9+c3WX3oSWl11uUVROVlD+SqVN2HgORHbTasxs+eFGsuj4R2c9IbWH/o0wze/9m95iJfcA+libQ9u/fjw996ENxy6+88krs38+PBtqpxbkD9/maIUd81EUroSN2PSOYvJFdjJxvo20iH1W5/agtGc30YQAAvhh4TaXPSWOPor34cTMBEj7t/B2eDSyOO2dfcL2GWmd2/H2IUk29TsX3mUQEyJAhAwn3IwMxfbEQrpOUHlq1+BbnHtXrPdseUWokugdi/6NsYPb+Tb1dR49rjBIgsw/YyNIEWnFxMXp7e3HppZdGLT916hScTuYS2MlogmPsehXCEC7ILgygFAAwGecwThjFabmcKTFZzMj5NpvqSamhdR4A9fSmueJxnjMqaIn6jEeaBhnKA3JlOBGEAGCqcBZAdGKUso1e3/uQ6MEfpXr2OUoJvVrM6z1R+qjdAwFMGaTsYvb+TSvJ/I9SfTj1NQAHBMhRKZxKqqwfjnBy+d1FL7MP2MjSbNcNN9yAe+65B7/+9a8xceJEAMDZs2fx3e9+FzfccIOtB0hAW+BKHJRqIEPAQakGm3zL8X3fZ8Ox6R8WD+M0yuGRquGAhCBEyBDgFvwolgPhQtIrl4dfrxd7WFCylNFEnyG5BF4U4aBUg43+G8PbUvponSutZTw/VGiUCPbI+jRVGMRkYRjvSdXYFliKFucObHFvNLQfZZsm8ShkCAhAxFFp2lgCrgABwHhcwGFpBg6MffVhMoZZ7yglYq/rbwTm4i7v6vDNzn2uX7LdEaWB1TGWWo2qF3vQ4tyRcH/Ktkp/bxKPol1qCP9sZB9UeJSxy3tSFR70fR5T/YMALk74xrYbeezT9UNyCXYGr0C/XIZLxZ6o9lYvnkCLcwcAhNuksiy2Dca2W7ZT8yxNoD322GO4/vrrMWvWLFx55ZUAgL1796Kqqgrbtm2z9QAL3X3e2/C6dEX4ZxkCujE1/LMXxVGvK0+Hioxwjo1zBrRj1Sn7qcUgn5Cn4F5fMx7m+SSiLKEVwR5Zk4zUotj9HJRqNZ8LIgM4h5KoZQMo4/WRUk6tnXKcRZS9tGqUHXWJ/Z9imR0TAYhaP3a92PYW+kWiqLovpQ2yTtnD0gTajBkz8Pbbb+OZZ57Bvn37MG7cOPzd3/0dPv/5z6OoqMjuYyxob0jzUrbvQop17vXmT7v8KT6i8YqIn/k+gjqfJ63Ho8inf2MiSp7RCPZEtcieKPfCqXeUGWrttJDGWUS5Rq+22FGX2P8pktkxkZlnw4bEPwMwtg2yTtnD8gPLxo8fj7vuusvOYyEV5juPOYUSaZtPqYKdDVVq10gAwHtSFR47OjO9B0REpMJMfdFb1646VSj1jjJDq32x3RFlp0R90466xP5PCrNjIrvmACLfl3XKHpYn0LZt24YtW7bg3Xffxe9//3vMmjULTzzxBC699FLcdNNNdh5jQQuljaVuEq1QIm2bZ55CldufeMUc8D304l3UqL52idiLb87uSvMRhfR6i/JqopKIkmMmgl2vFtkV5V4o9Y4yQ6udst0RZadEtcWOusT+TwqzY6LQs8/tHfuwTtnD0gTapk2bcP/99+PrX/861q9fj2AwCACYNGkSfvSjH3ECzUaLxA68Ll2O6Mja2Ahbq5G2hRPrXOX2o7ZkNNOHYYsvBl7Dfb4WlYlVCV9wvYZaZ378PYkot6lHsMdLVIvsiXIvnHpHmaHWTgtpnEWUa/RqlB11if2fIlkZE+mvHzsOksbyN7XbIOuUPfTPoIYf//jH+Jd/+Rfce++9cDovzsFdddVV+POf/2zbwRGw3v0MrhffgQAJoS90SqhFH9wYBSDDjVFcL76DRrELLvjhxihc8KNG6EON0IdxGA3/d+TrjWIX1rv4wMBctMi5H+td21Aj9EGABAESaoQ+PMzzSURZJHSt2q5bn4zUosj9KNvc5twZ3m8R/EC4RsooxXk4EURocCmjAoO8PlLKqbVTjrOIspdWjUq2LrH/kxqzY6LYNha7Xmx7e9i1Hetd23TbIOuUPSx9Au29994Lp29GKi4uxvnz55M+KIq23v0MgOioZSA05zyKInjkaWgtekm38W/xLcMLgYXwwgU3fKjEILYFluL7vs8xwjYHWY3rJiJKp8gH13ZKVbhU7ImqN0pde9D3eTggIQgR9THx7EqN2uLemMm/ChUQpV1Gtj8jNTeZ2mz1PYnInDcCc7HRfyNOypMBANOFATzgei6uLin3SGr1SOnr7KNklgwBImTUiac1r/Ox9eA+1y8Nt7VE67HdJs/SJ9AuueQS7N27N275yy+/jLlz5yZ7TKRCiZ09KNWO/b69aOx37SJOyFNwr68ZbwTU/+23+JbhmcBSeFEMQIAXxXhdugIHpVp44QpH2GptT0REZEVk7YqtN7F1zYti+FGEg1ItngksZY2ijNBrs/n0nkSFSOlrJ+QpkCGG76Pu87XE1SWlL7IekR2MXudZD7KfpU+grVu3DqtXr4bX64Usy3jrrbfw3HPP4ZFHHsG//uu/2n2MGdXrdWX6EAAAP8VHEqwh4me+j6DO54l75T9wbcL9yzrbp1q2/BsTEZG99CLTzQTkMGad0kWvzaaq/WXiPYkKkVpfA0KfCjJal9g3yQqj13nWg+xnaQLt7/7u7xAIBPCtb30LIyMjuPXWWzFjxgw8+eST+NznPmf3MWZEWVkZil1F2N5VnelDAQB0NlQl/Lzge1IVHjs6M2756GVFhp61rLV9OhS7ilBWVpaR9yYiotTQi0w3mzDNmHVKB702m0/vSVSI9PqUmbrEvklmGb3Osx5kP0sTaADw5S9/GV/+8pfx/vvvQ5IkTJ061c7jyrjKykr884+fwtDQUKYPBQDQ+mInDvV5ddeZXVWKNZ9YgyeffBJr1qxBTU0NAOATPz8Eb0BO+B6zq0qxYeUGW47XrLKyMlRWVmbkvYmIKDX0ItPNRrQzZp3SQa/N5tN7EhUirb6mvGa0LrFvkllGr/OsB9nP0gTa9773PTz44INwOByYMmVKePng4CBWrVqF5557zrYDzKTKysqsmdRZu3w8Vm1vh6wzD7Z2+TzUuEMhDjU1NaivrwcA/N21fmzapf/VTEEIbV9fnx2fuCMiotyXKDJdO6I9Op6dMeuULonabL68J1EhanHuwL2+ZsR+rUeArFOXWI8oeUav86wH2c/SBNq//du/4dVXX8UzzzwTnqTZuXMnbr/9dsyYMcPWA6SQZfOqsbm5CRt3enDgVOhTcf6ABAAocooIBCWsHJtgExtux7+8dRo/HDs3314xBwDwi993YsQXhMspYlyRA8NePwBg5uQSfPfGRnx8HifPUiUyQTVR0pze99uZ0kVEuWSRcz/WYzu2B5bAI02DOHb92+i/EYNyydjXZUK/GXIgCAdk1Iun8CHRgz9K9eFrXbNzZ9S1jtdCSpXINqvV/qxSa7cAsNF/IxDuCzJqhH7cXfQy2zSRSZF9rEII3S/1y2VRdeJhbMcm/wqckCsAADNi+lts/09Uj4iMWOTcj1ulXXghsBBeuOCGDx8WD4eTyAEgACcAGZNwDiXCaLjt6o2BtNo5pY6lCbS3334bK1euxIIFC/D444/j8OHDePLJJ/Gd73wHDzzwgN3HSGOWzavGsohJrraOHqzc1g7f2ESaQhKL8Py+AUwqPxiePPv2ijn49oo5qtscGxhB4i94klVKmorCP/b/B6XaqI/oKikr67FdM9I4cj+J1iciygbK9Sny+nVCnhK3XhAOfM65Eytdbbr747WQUm2Rc7/tbUmt3ap9EgYQ0D12Y09ExsX2scg6E1sn9Pp3Kvo/0RuBuXgmIiDAi2K8Ll2hsqaAAZRhQJbwsCt+XGOmnVNqWJpAmzhxIp5//nnce++9WLlyJZxOJ15++WV89KMftfv4SMfGHUd1X//F7zvDE2h628gysHGnJ2pyjuyjlfijRi9lhaksRJSrjF0HBbwQWJhwAo3XQspF6n1AKx2K7ZnIrER1hnUi9/V6XZk+BMt+io+Y3ELEz3wfQZ0v+jFMifYja2yXz9LdLiyHCPz4xz/GE088gc9//vNob2/H1772NTz77LOYP3++ncdHOg73ntN9fcQXNLzNkd5hW46J4plNTTGbvsJUFiLKdkavU14kHgTxWki5yK6xABGpM9Jn2K9yU1lZGYpdRdjelbsf9uhsqNL+nYmG96QqPHZ0pun9qG2X74pdRSgrK0vLe1maQFuxYgXeeust/Nu//Rs+/elP48KFC1i7di2uvvpqPPTQQ/jWt75l93GSitlVE7Cve1Dz9RKXw/A2l1WV2npsdJFe4o/W+mb2w1QWIsp2Rq+Dbvgs74vXQspmdo0FiEidkT7GfpWbKisr8c8/fgpDQ0OZPhTLWl/sxKE+r6ltZleVYsPKDab3o7ZdvisrK0tb+KOlCbRAIIA///nPmD59OgBg3Lhx2LRpE/7qr/4KX/rSlziBliatSxt0kznvXFhnaBtBAFYvqU/NQZJqmspFxpN9mMpCRLlK/zqokHGLc4+lffFaSNlOvQ9ICI0BhJi12Z6JzEpUZ1gncltlZWXaJkhSYe3y8br37erbzEN9fXXMsvFYta1d8/nlgqC+HdnH5AcJQ1599VV4PB40NzfjmmuuwYkTJwAAAwMD+NWvfmXrAZI2JZlzfm05nKIAcWz8JUp+fH7+ZHxr+RzdbUpcDsyvLceW5iYmcKbQIud+rHdtR6PYBRf8cGMULvjRKHbhNudONIpdGIdRNIpdWK/ysEi1/RhZn4goW8Rev2qEPpTiPJTUwSL4cZuBAAG1ffFaSLlArd0+7NqOh13bUCP0QYAEARJqhD7VB0cTkT61OlMj9LFOUFZQuwe/e0k95teWw+UU4XKKEACIAlA3ZTx+0qJ+f75sXjUevGEGSi6chtspoK6iBHVTxvO+Po0sfQLthRdeQEtLC2677Tb86U9/wujoKABgeHgYjzzyCBYtWmTrQRa6to4ebNxxFB0nhxCUZcgy4HKImFjixDlvELOrJuDp2z6EZfOq4fF4sG7dOnwp5mObyj4O957D7KoJaF3awNCANFISfZTYYY9UjfekKhyVpqFe7MF9rl8CCD0A9fu+z6FO7EWTeBTtUgM8UjUckBCEiHqxh/HERJRzIiPXlZh1AOFlRQjg2cBiPBNYAgEyFokdWO9+RnN/kSlpyr6VayevkWQntbYb2b4SvR67ToUwhMnCMN6TqrEtsBQtzh14dtxj6f5rEeUlowmaev02tr8CQL9cFrXeFt8yvBBYCC9ccMOHW5x7DP0CiArbsnnVSd9/P/ryQfz8dyfhdVfCDWDF5dPw7RVzwvf6a57fG3ev/+jLB7F1Tycu+IMYV+TAnQvr4oIGyThLE2jr16/H5s2bcfvtt+P5558PL1+4cCH+4R/+wbaDo9DE18pt7XHLfUEJfcOhZ8Xs6x7Equ3t2NzchAZ34n1Ers9JtPSJjR32j/1/KHK4BXLEVzgOSrVRz3GIXpfxxESUO2KvfQelWtzra0bkh+AjwwNkCHhduhz3eW/TnUTT2jevkWSXRO3LSPuLXeeEPEV3fSJKLb1+CyBhf10U6MDr0hXh5V4U45nAEgDgJBql1KMvH8SmXWPpmoIAb0DGpl0evPf+ebzS0RNeL/Jef2/X2YvbALjgD4Z/5iSaNZYm0A4dOoTrr78+bnlZWRnOnj2b7DFRhI07jhpaT5aBjTs9eHx5/ISY2j6U9TmBlj568dpy3PNPtDGGm4hyifq1L9ETJAS8Ic2ztG9eI9Mv3RHy6fJTfCRumQwRP/N9BHU+T8LXtfaht36+yde2QblLr24kGo/LEDVqk4AXAgs5gUYptXVPp+rytv09ccuUe/3DPcOq2/zi952cQLPI0gTatGnTcPToUdTV1UUt3717Ny699FI7jovGHO49Z3jdI73DAOInxLT2EVqf0sXO6GzGcBNRrrB6vTLyiwWtffMamR5lZWUodhVhe1d+/jKus6FKda73PakKjx2dmfB1vX1orZ+Pil1FKCsry/RhEAHQrxtG6o7WOpGfpCZKhQv+oOpyrWCCI73DmtuM+NSXU2KWJtBWrlyJNWvW4Gc/+xkEQcDJkyfx+9//Hn//93+P+++/3+5jLGizqyZgX/egoXUvqyo1tQ+t9Sk1zEbYJ9oXEVEusHrtEzQzphLvm9fI9KisrMQ///gpDA0NZfpQUqL1xU4c6vPGLZ9dVYoNKzdovu4ePYM1a9agpqZGcx21/eWrsrKynE7Po/yiVzdkCAnrlQBZdRLNDZ9tx0ikZlyRQ3VCTBDUJ9EuqyrF4R71SbQSlyMVh1gQLE2gfetb38Lg4CCWLl0Kr9eL66+/HsXFxfj7v/97fOUrX7H7GAEADz/8MP7f//t/2Lt3L1wuV8F8VbR1aYOhyFtBAFYvqQdw3tA+Lq5P6aIXrx0qxjKif00tIz7anjHcRJRb1K99EgQIOr/tDwUJWNk3r5HpVVlZmbeTI2uXj1cdP61dPg/19dXqrwOoHtiHmprrUV9fr7pOpMj9EVHqJaobWmN1Zb1FYgdely5H9Bhdxi3OPak7aCIAdy6si3qemWLZ3Gq07e9Rvdf/U8wz0CL3RdZYmkADQhNa9957L/bv3w9JkjB37lxMmDDBzmOL4vP58Ld/+7e45ppr8NOf/jRl75ONZk0uQdfACKSYwZcoAKIgoMgpIhiU8YOXDsDn96On4XZ8evsRDI8eQmBsozK3E5PHu3B6eBRTS4sBQDWlg1JnkXM/1mM7NvlX4IQ8ZeyzFTJqhH7cXfQyAGB7YEk4EehDogd/lOrhkaZBhAQJIurFU+ECf5d3tW7qF6VHonTByERVni8qVNOFAZyQK8I/FyGIa8SD6MNEdEpVcCKIYYwLvz4BI1jm/FP4Z73EtOnCAE7KkwEAkzEMGaEbIPhCr7UWvcT+Rqa0dfTgkZcOoGtgBJABpyggKMkQBGDKhGL84KUD4THUquvrsefdfhzpHcZlVaW4Zc54vPh0V3hfy+ZVY3NzEzbu9OBI7zCmlhbjvC+A/nOhT6vMnFwS/qwlE9OJkhNbK6biLN6SZselZa7H9qgxd7NzZ7hOrMd2PO67Cf2I/Npx6Nc904UB1Irvo0IaRj9Kw685IKNdasAbgeOsN2RYomt+5OtVZaH7d6UeybKMIoeISeOLsOtwH2ZNLgEAnB4exWVVpVhYX4Gnx7atLHVh8EIAvoAEQQAcgoDfHX0fbR09rDEWWJ5AA4CSkhJcddVVdh2LroceeggAsHXr1rS8XzbQSuBUSDIgyTICY99h7uwfCb0gFuHMheiPag55AxjyBnD34vqoWWgmcqZftxz5m3ohfFNpJnqbqXPZQf1c6Ceq8nxRIYntIwo/ivC6dDluc+7EFvfGuPXOYbxmKprSh26VduGZmIdB92Ni1M8n5Cm4z9eC9djG/kaGqI29lF9GynLo5kSxr3sQb58YjBpDeTwevBizz2XzqsOvx+6/s38Eq7a3Y9X1HJ8RJUNtTHYQF8dfsWmZWjVhv1QbV0uUhwqckKfE1R1AQBAc35E5sbUg9pqvViti+SUZp4d94dcFAdjc3AQAUdte8AfDdyayDARkmTUmCUlNoGW70dFRjI5eHOjk2jM6jCZwmrH1951xy5jImT52JMYxdS57qJ8LIw+g5fnKF6msM/mQXqefQCjgPwLX4sbATt00Q7VvvskQ8R+Baw0dgwwh71IO86FtZCuzYy+zYyitZPRfqKSrcXxGily/p0kHvbT7ixKnZb4QWGj5GDi+I6O0aoFyzU9Yi4T4+w1le7VnBqiOpVhjLMnrCbRHHnkk/Mm1XGQmgdOoCxqJG0zkTA87EuOYOpc9kvk35/nKD6moM/mUbJgogXBULsJjR/TTDAGovjYqF6k9JlJVPqYcMtkwNayMvcyMobT2P6KRlMbxGQG5f0+TDkbHVYnSMpNN0+T4jozQqgXKNd/qPMCR3uGEz05Xez8yLqMTaA8++GDCYvB///d/lr8mes8992Dt2rXhn4eGhlBba08KYjqYSeA0apzLoTqJxkTO9LAjMY6pc9kjmWRVnq/8kIo6k0/JhokSCN1FIjZs0E4znF1VChlQTzosEuENGBsl5mPKIZMNU8PK2MvMGEpr/yVFDtVJNI7PCMj9e5p0MDomS5SW6YYPXhQndRxEiWjVAuWab3Ue4LKqUmDsK5qG1ydTMjqB9pWvfAWf+9zndNepq6uzvP/i4mIUF1u/AGaa0QROM+68pg6bX/cwkTND7EiMY+pc9lA/F4kTVXm+8keq6ky+JBsmSiD8wnWXaiYVKumEMoBV29qjvn4gCMDfXXtpXD1TI4Aph2Rc69KGuPamx+wYSisZ/Y5r67B5F8dnpC7X72nSQT3xOTbRPnFa5i3OPWPPStNOidZ6jeM7MkqrFijX/ITzALIMQRDixkarl9SHxk0qCdGxu2KNsSajE2hTpkzBlClTMnkIWU0tuWnEF8D7SnJTxXismFcdTn+aWloMv9+PU2dHUD7ejaFRKfzg26mlxVh/8+X4+LxqLJhZHt7nZVWlWL2kHh/nd5/TQkni1Er+Sdc+yB5a5wJQT1Tl+aJCE1nHOk4MQpJlSDJQ4nLgzoV1+NbyOXHrqdWmB2+YgUd/8ydIpVX4wLSJ4ddi69nC+gq8/OdTofREhBIOv3tjI2scGbZsXjU2tzRdTOEEMGWCCyUu58Ukc0HA6SGvpTGUXltfUMvxGZFVamOySgyqpnDqUV5/IbAQXrhQhADKMIIRuOPGdBVC6JPi/XIZx3dkSqJxj9o8ABAKspk5sQi+vf+FO+68E/958Lzq9mr7lgHWGBvkzDPQurq6MDAwgK6uLgSDQezduxcA0NDQgAkTJmT24FIoMrlJjxJze2jwAty+s1izbAHu+OgCPPryQfx093s4PTyKu7a1o7LUhfU3X4Ffrzb28GVKDRlC+I8VRhM7KfW0zkXksjcCc9EuNSR1zolylVLDHnnpAI4NjEAUQr/UkWXgpqd2h+PbF9ZPAWQ59BtTWQ7/prStowfP7u2Ht3gSLptUjNaYAWZsjfz22KScsu3TO45izfN7VSPiidQYHXvZvf9Uvy9Rvks0Pn4jMBd3eVfDI1XDAQlBiKgXe9Di3BG13UpXW9RE2xuBudgWWIr3pGrIEOLWJ7Ii0TVf6/Vf/HYvHj2+AOt/exJOh4igJEeNmyK3VeYIOA6yj86jfbPL/fffjyuvvBIPPPAAzp07hyuvvBJXXnkl/vCHP2T60DJOibnd1z0Ib0DGiLsSD756Aqu2tWPTLg98QSm8bt+wDyu3taOtoyeDR1y4lIjtg1ItvHCFI6/fCMzN9KFRivCcU6FTalRn/whkGZDkUNz6pl0e7OsexAV/EPu6B+N+XrW9HY++fBArt7XjUJ8XkliEQ31erNpurIZF1sbIfbL+EREVnsjxmB9F8KIYfhQlHJdxHEfZpK2jBw+8egIj7kr4JRkX/EH4gpLqGIfjoNTImU+gbd26FVu3bs30YWQl1RhcAG37tTsHI2szQy1im5HX+Y3nnApdwih2DbIMbP19p+pyIzUsUUQ8EREBvd7kUidzxU/xEc3XZIj4me8jqPN5DG2nt36+KJR2kWv0xlSxYxyOg1IjZybQSJtWzK3eg5UZWZsZWtHWjLzOXzznVOisRrEDUE2NBozVsEQR8UREhaysrAzFriJs7yqMG+nOhird7169J1XhsaMzDW+ntX4+KXYVoaysLNOHQRESjakixzgcB6UGJ9DygFbMrSBoT6IxsjYztCK2GXmdv3jOqdBZjWIHgHEuh+okmpEaliginoiokFVWVuKff/wUhoaGMn0oadH6YicO9Xk1X6+bFHpI+5o1a1BTU5Nwu9lVpdiwcoP9B5pFysrK8iIRPJ8kGlNFjnE4DkoNTqDlAdUYXADL5lbjFY3vODOyNjPUIrYZeZ3feM6p0LUubcDKbe2mtxME4M5r6rD5dY9mzHui99WLiCciKnSVlZUFM0Gydvn4uJqgEATgzg9Pw4v/B9TU1KC+vl53O0EA1i6fh/r6wvj0HmWP1qUNWLWtHWqfkYkd43AclBo5EyJA2pSY2/m15XA7BZRcOI2HbpiBzS1NuHtxPVzOi6d5amkxftLSxMjaDFnk3I/1ru1oFLswDqNoFLuw3rWdz8LKYzznVOiWzavGlpYm1FWUQBQAUQDqKkpw95J6zK8tR4nLgfm15bh7cfTPW5qb8O0Vc7C5uQlzKt0QJT/mVLqxpdlYDYusjZH7ZP0jIio8kTWh2CmixOWAyymGa8O1deqfymEtoWyybF41HrxhBkounIbLIcS148h2ybabGvwEWg5Q4mcP956Li59t6+jBIy8dwLGBEUAGnKKA0eLJeGZvP6qqe7BgZjkaPaVR27LTZJZWxLYSkd0pVaFO7EWTeBQ7g1fgpDwZADBdGEBr0UtY5NwfXjdRDDdlh0Sx6kT5LjaKPbauLby0Ans87+PAqWE4RAEHTg7h6R1Hw79hjfx/GRdrX9fACGQAsyaX4J4bG6Mi29VqJpEdIttYVVnoa1+9Q6OYXTUBjRUOHJz5SXzi54fwgepTbH9EaaR1/Y9d/s+fvzKuX3o8noT7AkIPZl/z/F7d/bPfkxarbSX2nt/lKMa9H5mOOz66wNA2AoDBEZ/qJ9fIHEGW9R41n1+GhoYwceJEDA4O5swDEZX42UiCAGxubgIA01+LUbYtlIu6x+PBunXr8M3ZXagtGc304WhSIrITESDjVudOPKOS7Bh6XcrYp5uOjxTjscMzsWHDhqiPvtshV85jPjB7HnPxuppK/PdITK2uaREAU4O9uxfXY9Ou6JugQqt7lFpm2i/A9mcHXlfj8d8kntY906rrjdUFZay5YcMGHPWOj98X4uuRmf0T6d3X67UVrbojANjcor6tlW0KmZlrKr/CmeX04mf1Ymy1KNtSdtmmMSEWS4aAFwILdV4XsT2wxKajIiKyn5naZfY3fL/Y0xm/D9Y9spHZsRfbH1F6aN0zbf19p+pyvX6pui+V9azunwqT3n292e2AUJvU2tbKNmQMv8KZ5fTiZ61+dpDRtdmnU6oyvK4XLtv2RUSUboki2JMx4o9P7ARY98g+Vtov2x9R6mn1TbUkZ0C/X5rp51b2T4VJ777eynZ621rZhozhJ9Cy3OyqCarLL6sq1XwtEUbXZp86sdfwum74bNsXEVG6Wa1dRpQUOVSXs+6RXay0X7Y/otTT6pvjXObrgpl+bmX/VJj07uutbKe3rZVtyBhOoGW51qUNEIToZUr8rPIwSz1C7M+Mrs1KLc4dECDFLI3/iKEAGbc496isq7wuodm50/4DJCKyiVpd02JwtbA7rq3TrJlEdjDTfgG2P6J00bpnuvMa83VBdV8q61ndPxUmvfv6RNupEaC9rZVtyBhOoGU5JX62rqIEohBq9EWiiK88+yds3HEUdy+uD78WNvbdzqmlxVgV8boghJLKCiY1Igu9EZiLu7yr8fGRf8Bd3tV4IzAXQCilcb1rOxrFLozDKBrFLtzm3IkaoQ8CJAiQUCP0Yb1rG1a62sLruuCHG6NwwY9GsStjAQJERGbMmhyqS6IA1FWU4O4l9ZhfWw6XU0SJy4FiJZK9pQlbWpowp9INUfJjTqUbPxlbFlkXXQ4RLoeIPUffx6rrWfcodZRx2fzacpS4HKirKMHUUle4PU8e54Br9CzcTgF1FSWYNbkEa57fi5ue2o1HXz6Im57ajcbvvYKbntqNto6eTP91iPJGbN+cX1uOLc1N+PaKOarLP67zEHXVfbU04e7F9eFPOpcUObBqcX14/6w7lIhWG9Vri8p2W1qaUFl68TE+ghTA5+ZP1txW2SZyrCQKoY9n3LWtHbPvfRmPvnzQxr9d4eAz0HJEZ/9I+L99wdCnj/Z1D+LtE4PxiZxjU9unh0fjUmE6+0ewant7wSXD9Hr1nxuWDn/AHDyBz4d/PijV4j5fM77u+yWuwkHUwYP7MHa+pNCfG7Hz4g5kAD7guK84el2FdPH1TMiGf2Miym5qqVDHBkawoLYc314+R3O7Bvf5UDrayg2orw/VrmXzqsP7i6qL3YNRNy6FWvcodZbNqw63pdg2PXAhCLgm4nPzJuH5fQPh5fu6B7GvezDqZ7ZLIntF9k0jy83sq62jJ+q+asQfxOZdHiyoLQcQfa/GukNarLRFRd/wxcf4yKITz+8bwJIP9mjuT3kvpU5FPj/dF5TC7fnbK7THXxSPE2g5QC/xKZzcYSJRQNmmEC7oZWVlKHYVYXtX5v+uB2d+DHBHL5MhYsuFj2LX8RH1jXJMsauIcepEpEkvgcpKTTKTlFYodY/SS3WMJgh4seNMwm3ZLolyh26Cosp9GPs32UlrvGOkjenNJfzi952cQDOJE2g5IFESjJVEzkJJ36isrMQ///gpDA0NZfpQ8ImfHwIC8SdKKq3Chg0bopZ1d3fjySefxJo1a1BTU5OuQ0xaWVkZKisrM30YRJSlrCZQmd2fne9BpEerDXpV6r0atkui3KBXv7Tuw9i/yS7JjJ/0xkojGimypI0TaDlgdtWEqI/9x7qsqhSQZd11VLcpEJWVlVkxqfOB6lOq5+gD0yaivl79YY41NTWarxER5Rqtema1JiWqj3a8B5EerTbodgqGJtHYLolyg2790rgPY/8muyQzftIbK5VopMiSNoYI5AC9xCcziZyx21B6WU1eISLKF3ZfB80kpfFaS6mgOkaTZXxq3qSEaZ1sl0S5Q69+cYxPqaY13jHSxvTmCe5cWJfcgRUgTqDlgMjEjriEsrHkDiVpY0ZZESBLEBBKNlPSysymfZD9rCavEBHlC7uvg1pJaax7lC6xbXBOpRuXnvwffOnDU+Pa5t2L69kuiXKUXv3iGJ9SLbKNuZ0CSi6cxkM3zDDUxtRSPF1OEa1L6vEtnQAnUsevcOYItcSOto4ePL3jKNY8vxezqyagdWkD/u2z9aGksg0bor76pyRwbIxZnw+2TB/l3//AqWE4RAEHTg7h6R1H8aeus9jjeR+He89hdtUELKyfgtf2n8ChhjvxsX85CBkHMa7IgTsX1mHBzHJs3HE0vC7PYXZ5IzAX2wJL0SlVoU7sRYtzBwDELVvk3J/hIyVKH+XaF3ndal1SH1rWM4yndxwNP/hf7frW1tGDx1/pxJGG29H6YifWLh8fVdMOnAo9/8MXlLDv+Fncta0ddRUluOfGRl4fyTZq7TiqfckyZBkYHg2ir/LD+MTPD+ED1WWs00QZkLC/qtjdOYy3L/kcPvovBwEchMsh4ovXXYIFM8ux7t/3YcgbABD61M+yedXY3NKk/X5L6jl5RqYYabPKfIDH48G6df+Ka+uuN73f9TdfwZqUJEGWzT5+PncNDQ1h4sSJGBwczPmkwNjYdCD0UeEHPzYDv376H+Im0LTWZ7xyeqj9+9uhUM5hqFCswzdnd6G2ZDTTh6PqjcBc3Ou7PWqZABlyzBfKBEhY79qetZNox0eK8djhmXHXEC35dF21A/89oqnWHqgnZcYSBGDV9fXhmPVEy9Vsacn/6yOlnt4YCoBufS+UOp1KvK7G47+JNiv3PFbG6cvHJtF4j0XJMtuGlPuiRGN1tk3jzFxT+RXOHKUVpfzc3n5T62/cmfgGhJKnFx+cDJ7D7LEtsDRuWezkWWiZiO2BJWk4IqLM04pdN0KWgV/s6VRdvvX38ctV35/XR7KB3hgqUX1nnSZKLyv3PFbG6f+9v8fy+xFFSlUbYttMDX6FM0dpxdF2nhnFXBPrM145PfTig5PFc5gdOqWqlKxLlMuSvfaN+NXj1S8YjF3n9ZHsoDeGMvI9DrZDovSxcs9jpVZJsvX3I4qUqjbEtpka/ARajppdNUF1ed2kYlPrM145PbT+/e3Ac5gd6sTelKxLlMuSvfaVFKnHq48zGLvO6yPZQW8MZaSNsx0SpY+Vex4rtUoUrL8fUaRUtSG2zdTgJ9ByVOvSBqza3h71m09BAG5dUIEXf2d8fcYrp4fav78dCu0c9npdiVfKkBXYjUP4LOSo30vIY38uLhMgYbm0G8dH1Ce7My2b/40p96jWHqh/jTN2uSAAd1xbh827PHG1685r6rD5dU/Ca2ohXR8pdfTGUDKgW98LrU4TZZqVe57WpQ2mn4GmPEOK91iUrFS1IbbN1OAEWo5aNq8aq66vxy/2dGLEH4QgAA5BwJb/PY2+uk+PpT+dCid4KNG3G3d6cKR3GJdVlWI1E2LSJvLf/+CpIThEAQFJRuO0Miy8tAJ73u3HgbHl/oAEpwh4fQEIogMygBJXKIVTeSbQiD+IkiIH7ri2riDOYVlZGYpdRdjelc1/1xFMrfgz+ibNhSQ4IcoBVJ7Zj/HePvRM/iC8xZPgHj2D6oF92HX+AnZhJgDg7PhZ6KmYD6+rHG7fWVT370P5+WMZ/ZsUu4r4UGKyhXLte+SlA+gaGIEMYMoEFwQIeP9cKBCkYkJo0rZv2AcgNJE2q6IE372xER+fV40FteV44pUOHO4dxuyqUqxdPi+0fGY5Nu704MCpIQCAPyCFJ+BcThGQEU745MNyySitJDS9MVTka5PcIvr6+iCOn4QPTJtoaKxlJTGQiKJF9qNZk0sAAKeHR3FZVSkWXlqBp3ccxZrn96KqLPQLzN6hUcyumoCF9VOwx/M+HCIQDMqhGQYATlHAXddfivm18Smcyy+vxqaxEBHlnmzr7ztxwRfEOJcDd15TGONzsoeR+/TI9j2zvAj+8bNs2S+ZxxTOHGU0LYZJG7lB9XzKMh76eA3u+OgCzXUK6fz29fVhaGgo04ehaXfnMB549UTUMgHAJSf+Bw986WbU1NQY3ubBG2bgurrMfby6rKwMlZWVhtbNp+uqHfjvEc9KupkAYHNEgiYTpygd7Gg/Rtuqne+Z73hdjcd/k2jJJOVqMdoP2Ycp1YzcJ1JyzFxT+Qm0HGU0LUZJ2uAFPLupnk9BwHN7+3HHR7XXKaTzW1lZaXhSJxPWvrw7bpkMoGfyB1FTU6N6I6W1zX8ePM+CSHnDSrqZDGvXtkK/TlJyMtF+2GaJkqebNmjxsyJG+yH7MKWakftESh9OoOUoM2kxTNrIfnqpqonW4fnNDlrnx1s8yfQ2PKeUT6wmcVrpB+xTlIxMtB+2WaLkJZuUq8VIP2QfplQzcp9I6cMUzhxlJi2GSRvZz0iqKpNUspvW+XGPnjG9Dc8p5ROrSZxW+gH7FCUjE+2HbZYoeckm5Wox0g/ZhynVjNwnUvpwAi1HtS5tUJ5xqYtJG7lB9XzKMm5dUKG7Ds9v9lA9PwCqB/aZ24bnlPKM0XoVSYC1fsA+RcnIRPthmyVKnl4/slKDIrdP5r2J7GDkPpHShxNoOUpJ1ZhfW45ip4gSlwMup4gZZUVwjZ6F2ylgfm05tjQ3MWkjB0SezxKXA3Mq3bj05P/g2ogHyceuw/ObXdTOz0M3zED5+S5T2/CcUr6Jbed1FSWoqygJt/m7l9SjrqIEogCIAlBXUYItLdb6AfsUJSMT7Ydtlih5ev1ItQZNGX+xBi2ux/zacrgcAkTJjyIRpvoh+zClmpH7REofPgMthy0bKwqRQulPm3HT6vvx8z+ewcrt7RAAzJxcgntubAyvz8j07BN5Pj0eD9b9Ln7iRe2cU/aIPT8ejwcvmtyGKB9FtnOl/vQOjQKyjAW15fj28jlR67d19OCmp3aHa9TfNJr8Co4sh557I8somKhxskUmrsmsA0TJ0+pHRu95lATdm1bfj/88cA5rnt9r+B6JfZjsotVejdwnUnrwE2h56Oz4WXjg1RPo7B+BLAOSDHT2j2DVtna0dfSEo3D3dQ/igj+Ifd2DWLU99BoREVGqGKk/aus8+OoJnB0/y5b9ExFRYTBbE5R7KNYQygSOYXIDJ9DyUE/FfNXlMkKRyrpRz0RERClipP6orgOgZ/IHbdk/EREVBrM1Qe0eijWE0oVjmNzAr3DmIa+rXPM1vThnxi0TEVEqaUWxR9YfrXW8xZNs2T8RERUGszVB6x6KNYTSgWOY3MBPoOUht++s5mt6cc6MWyYiolQyUn+01nGPnrFl/0REVBjM1gSteyjWEEoHjmFyAyfQ8lB1/z7V5QK045wZt0xERKlmpP6orgOgekC9tpndPxERFQazNaG6fx9iVmcNobThGCY3cAItD5WfP4aHbpiBuooSiAIgCkBdRQm2tKjHOTNumYiI0sFI/VFb56EbZqD8fOLEKdY3IiJSmK0J5eeP4cEbZrCGUEZwDJMb+Ay0HKAWZwsAj7x0AMcGRiAAmDm5BPfc2IienmEcnPlJ7N9xEh+oLsOm5ibVWGXGLWevto4ePP5KJ4403I7WFzuxdvl4nqsctbsz1B8/8fND+ED1KUNR6ET5SKljB04NwyEKCEoyGqeVonVJverAUKlRynbf/+0JBBtux8f/9SBcziMXt1fpU6xvlG3UxnFso0T20upnkTWhraMHT+84ijXP79Xvi7Icema0LEPj0dFEKZFoDKN3n6hXa1iH7CPIstYj5fPP0NAQJk6ciMHBQZSVlWX6cAxR4mwjCYDhi7kgAJs1JtEo+6ieb57DnFQo5zIXr6upxH+PeGp9QaHXJ/S2M7I9USZ4PB6sW7cOGzZsQH19fcHUglTidTUe/02iGelnidbxeDz48oNP4d0ZH9PdD1Gm6LVhAJZeY7sOMXNN5Vc4s5xqnK2J7Rl9m1sYX5w/eC6JQtT6gkKvT+htZ2R7omzAWkCUekb6mZF1eirmJ1yHKFP02rDV18g8foUzy2nF2ZrB6Nvcwfji/MFzSRSSqI5p9Qmj9Y99irIZawFR6hnpZ0bW8brKE65DlCl6bVjrO4WJXiPz+Am0LKcVZ2sGo29zB+OL8wfPJVFIojqm1SeM1j/2KcpmrAVEqWeknxlZx+07m3AdokzRa8NWXyPzOIGW5VTjbHXWZ/RybmN8cf5oXdrA/kgE9euaQq9P6G1nZHuibMC6TpR6RvqZkXWq+/dx7EZZS68NW32NzOMEWpZTjbNtacKWlibUVZRAFABRACpLXairKIFTBETJD5dDCEffygBuemo3Gr/3Cm56ajfaOnoy/dciDcr5nlPphij5MafSzfjiHLVsXjUevGEGSi6chtspMIqaClZkHSt2iihxOeByign7ROR2RWO1zSkCJS4Hig1sT5RObR09uOmp3fjEzw/h4MxPYndn6KsxquO4mHarbMtxGpE1RvqZkXXKzx/DgzfM0F0nEvsupZPefaJe+1ZeU+YOBAGYNbmECbMWMYUzD2gldW5uYepGropN8aLcVAjnMV+vq1bx3yM1CqEvUe7SG4clGmsxpTMxXlfj8d/EfmbrDPsuZYqVMRHbqz6mcBYYraROpm4QERERpZbeOMzSthynEWU99l3KJWyv9mEKZx6wmshBRERERMlJJmmTKZ1EuYl9l3IJ26t9+Am0PMDUDSIiIqLMSGasxXEaUW5i36VcwvZqH06g5QGtpE6mbhARERGllt44zNK2HKcRZT32XcolbK/24QRaHohM3XA7BZRcOI2HbpiRMJGDiIiIiJKjNw4zsy3HaUS5g32Xcgnbq334DLQ80NbRg407juJw7znMmlQM/7G3cW3d9eHXl41NpBGRfSL73eyqCWhd2sB+RpQl2jp68MhLB9A1MAIZobj2e25sZB+llFHGWqF0tH+NGocZ3ZaIzMvkeIx9l3KJ0lY37jiKwz3DeHrHUcgRy8kYfgItxymRtPu6B3HBH8ShPi/enf5R7O7kAwGJUiW23+3rHsSq7e1o6+jJ9KERFTylf3b2j0CSQylTnf0jWLmNfZSIKJ9wPEZkHPuLPTiBluPUImkhCHhub3/6D4aoQDAKmih7qdZF5TX2USKivMHxGJFx7C/24ARajtOKpO08M5rmIyEqHIyCJspeWv0TYB8lIsonHI8RGcf+Yg9OoOU4rUjauknFaT4SosLBKGii7KXVPwH2USKifMLxGJFx7C/2yIkJtM7OTnzxi1/EJZdcgnHjxqG+vh4PPPAAfD5fpg8t49QiaSHLuHVBRUaOh6gQMAqaKHu1Lm1AbFlUsI8SEeUPjseIjGN/sUdOpHAePHgQkiRhy5YtaGhowDvvvIMvf/nLOH/+PP7pn/4p04eXUUok7cadHhzpHcbMiUXw7f0vXFv31UwfGiXp7PhZaH2xE11nj2B21QQsrJ+CPZ73mfqYBWL73WVVpVi9pF41ClrtPL78zimmA1LBs5qcprYdADzy0gEcGxgBZMAhCghIMgBAADBrynh8d8UcxrUTEeURtfHYwvoKPL3jKNY8vzep8bJSaw6cGoZDFBCUZEwvdwMAeodGORannBE5bpo1uQQAcHp4VPf+hbTlxATa8uXLsXz58vDPl156KQ4dOoRNmzYV/AQaEB2h7PF4sO53XRk+IkrW7s5hvDvjY0CfFwCwr3sQ+7oHw68rqSmbm5tYuDPESHR5ovMIXEwH3NLCc0mFQ0mCUhi9pu3uHMYDrx6M3m5bO+SY9ZTJMwCAANzDyTMiorwUOR6zWltixe4HwdD/dfaPJL1vonSKbcud/SMQBLDdJiEnvsKpZnBwEJMnT9ZdZ3R0FENDQ1F/iHLBswZSVJmakv2MnEcFz2VuYp2xxmoSlFqfip08s7JfIqJsxlpjjF0pg3ppzsnumyidmLxpv5ycQPN4PPjxj3+MVatW6a73yCOPYOLEieE/tbW1aTpCouQcM5iiytSU7Gb0PAI8l7mKdcYaq0lQZvqUmf0SEWUz1hpj7EoZ1EtzTnbfROnE5E37ZXQC7cEHH4QgCLp//vCHP0Rtc/LkSSxfvhx/+7d/iy996Uu6+7/nnnswODgY/nP8+PFU/nWIbDPLYIoqU1Oym9HzCPBc5irWGWusJkGZ6VNm9ktElM1Ya4yxK2VQL8052X0TpROTN+2X0Qm0r3zlKzhw4IDun8svvzy8/smTJ7F06VJcc801+MlPfpJw/8XFxSgrK4v6Q5QLbl1QEfp8rQ6mpmQ/I+dRwXOZm1hnrLGaBHXrgor47RK8F6+VRJTrWGuMsStlUG0/alhfKNsxedN+GQ0RmDJlCqZMmWJo3RMnTmDp0qVoamrCz3/+c4hiTn77lMiQ6+pKcenJ38K14K/QNegPpwrt8fQnTH2k7KF6Hi+tCKdwAsDMCqYDUuExk2Qb6bq6UtXtZIRSOLsGRiDLQJEzNEZonFbGayURUYGwWlv09nPw1FA42Xn6xFAKJxMMKVfY1SfoopxI4Tx58iSWLFmCmTNn4p/+6Z/Q19cXfq26mic/1tnxs9D6Yie6zh5hxHIOKz9/DBturkN9PX9DkMvUzuO3V8zJ4BERZQcjSbaaZDn04U5ZhpzsvoiIKG9YrQe7O4dxcOYn8YmfH8IHqk+hdWkDfr362hQcIVF6cYxkr5z4GNd///d/4+jRo3jttddQU1ODadOmhf9QtN2dw3h3xsdwqM+LC/5gOGK5raMn04dGRESUlN2dw1i5rR37ugdZ44iIyBZtHT144NUTGHFXwhuQWVuISFNOTKDdeeedkGVZ9Q9Fe3Zvf9wyRtUSEVE+YI0jIiK7bdxxNG4ZawsRqcmJCTQy7tiZUdXljKolIqJcxxpHRER2O9x7TnU5awsRxeIEWp6ZNalYdTmjaomIKNexxhERkd1mV01QXc7aQkSxOIGWZ25dUBH6zHEERtUSEVE+uHVBBePYiYjIVq1LGxBTWlhbiEgVJ9DyzHV1pbj05G8xp9KNEpcD82vLsaW5iVG1RESU866rK8Xm5ibMry1njSMiIlssm1eNB2+YgZILp+F2CqwtRKTJmekDIPuVnz+GDTfX4ah3PDbuOIo1z+/F7KoJaF3awAjbLNbW0YONO47icO85zCwvgn/8rEwfEhFR1lGLY4+8frLeUaa0dfTg8Vc6caThdrS+2Im1y8ezHRJlEb1acV1dKX59/L+wYcMG1Nfzk2eUGzj+ST9+Ai1P7e4cxspt7djXPYgL/iDjmLNcW0dP1Pk61OfFu9M/it2dfHgpEZGe2Osn6x1lgtIOD/V5IYlFONTnZTskyiKsFZRveL+fGZxAy1PP7u2PW8Y45uylFp8NQcBzKueRiIguUrt+st5RurEdEmU39lHKN7zfzwxOoOWpY2dGVZczjjk7acVnd2qcRyIiCtG6frLeUTqxHRJlN/ZRyje8388MTqDlqVmTilWXM445O2nFZ9dpnEciIgrRun6y3lE6sR0SZTf2Uco3vN/PDE6g5albF1RAiMljZhxz9mpd2hB3viDLuHVBRUaOh4goV6hdP1nvKN3YDomyG/so5Rve72cGJ9Dy1HV1pdjc3IT5teUocTkYx5zlls2rjjpfcyrduPTk/+DauujfILR19OCmp3aj8Xuv4KandvMhkUSUd8xe52Kvn6x3lAlKO5xT6YYo+TGn0s12SJRlZk0ugSgAogDUVZSwj1JO4/1+ZjgzfQCUOsvmVTPGNodEni+Px4N1v+uKel1JD1IoSSubm5t4nokoL2hd5x782Azd7VjvKBssm1eNBvd5rFu3DhtWbkB9PdskUTaIrS0AcGxgBHKGjofILhz/pB8/gUaUI5geRET5Tus6x0RiIiKyimNoIrILJ9CIcgTTg4go3zGRmIiI7MYxNBHZhRNoRDmC6UFElO+YSExERHbjGJqI7MIJNKIcwfQgIsp3Wtc5JhITEZFVHEMTkV04gUaUI5g0R0T5Tus6F5tITEREZBTH0ERkF6Zw5qGz42eh9cVOdJ09gtlVE9C6tIHpHHkiMmmlraMHT+84ijXP7+V5JqKs1NbRg407juJw7znMrpqAhfVTsMfzfvhnteuWWqKUx8MHPRMRkXVKXdm44ygO9wzj6R1HIUcsjxVbvzjOJiKAn0DLO7s7h/HujI/hUJ8XF/xB7OsexKrt7Wjr6Mn0oZGNlDjufd2DPM9ElJXUrlObdnl43SIiorQzM3bmOJuItPATaBkwOjqK7u7ulOx761un4pbJMvDEKx1ocJ+35T1qampQXMwHOmeSXhw3fztGRNlA7ToVi9ctIiJKBzNjZ46zif7/9u4/rsb7/x/44/TDKZ1KijJFKSUqP8LIj2qLY6YZsi2F3ppfmR+Z/HibMG8zGjEbNj+KtzYaZu/xJjS8EfIrGxrW1tJX5EcjB6nO9f3D55w5Th2lOr963G+3bpzrus51nle9rtfr6tnrup5UGSbQdCA/Px9xcXF1su8/PUYAJuZqy6/cLK61z0xISIC7Ox+6qUssx01E+q6yfup57LeIiKiuVefamdfZRFQZJtB0wNnZGQkJCXWy75idubh867Hack9HaySMrZ3PdHZ2rpX90MvzdJTgfP49teUsx01E+qKyfup57LeIiKiuVefamdfZRFQZJtB0QCwW19kMrqn9rDBu8xkIwt/LRCJgar92cHfnlGNjERPsUeHPmeW4iUhfVNRPPY/9FhERaUN1rp15nU1ElWECzcgoyjSvOpSDqzeL0drRGhOC3Fmm2cApKgFlFxTD1ESEcrmAlo0bAgAKi0v4cyYivVPReBTgbo+MnDu1Nj5pqpLGCmpERKSg6XektIs3sGxvLq56jEDMzlxM7deuWr9Pcbwhqj+YQDNC0nZO7LSNiKISkFL5039y7zyESASsifTnz5uI9FJdjkfP942KKmlrIv0BoNJ17C+JiOqnisYklbHExByXbz1Wjhc/TOjxwn1qGos43hAZHxNdB0BEmmmqZKeoCEREVN9oqpKmaR0REZFCTccLjjdE9QtnoBHpuRdVsmNFICKqjzRVSavsuWvsL4mI6Fk1rbjJip1E9QtnoBHpOU9Hicb1rAhERPVRZX1ja0drjeuIiIgUajpecLwhql+YQCPSczHBHhCJKl7HikBEVF9V1Dcq+kRN64iIiBRqOl5wvCGqX5hAI9JziqpB7V0aQWxmgoYNTNHAzATtXRrhq0h/Vt4konrp2b6xYQNTlT5R0zoiIiIFxXjRpokFTOSlaNPEolrjBccbovqFz0AjMgCsrGo8WOqcSFVNzglNfSP7TSIiqgppOyd4WMgQFxeHhLEJcHev3tjB8Yb0BX/PqHucgUZEpCWKUufn8+/hUWm5stR52sUbug6NSCd4ThARERHVHK+ptIMz0IioXispKUF+fn6d7FuxX8W/y/bmqm0jCEDi3ovwsJC91Gc4OztDLBa/dIxEurTq4G9qywQBWHUoh38xJSIiIqoiXlNpBxNoRFSv5efnIy4urk4/Y8WKFQCAqx4jABNztfVXbha/dAwJCQlwd+eDaskwXbn5oMLlV28WazkSIiIiIsPFayrtYAKNiOo1Z2dnJCQkaOWzYnbm4vKtx2rLPR2tkTD25WJwdnauaVhEOuPpKMH5/Htqy1s7WusgGiIiIiLDxGsq7WACzYiUl5cjOzsbRUVFsLOzg7e3N0xNTXUdFpFeE4vFWpvBNbWfFcZtPgNB+HuZSARM7deu2g+sJTIGMcEeFZ4TE4I4q5KIiIioqnhNpR1MoBmJEydOIDk5GYWFhcplTZs2RVRUFLp166bDyIhIQVHqfNWhHFy9WYzWjtaYEOTOUudUb/GcICIiIqo5XlNpBxNoRuDEiRNISEiAv78/YmNj0aJFC+Tl5WH79u2YvWorcOgvXJeBpWyJ9ABLnROp4jlB+iDt4g2sOvgbrtx8wOslIiLSmZqMR7ymqnsmug6Aaqa8vBzJycnw9/fHzJkz4eXlBUtLS3h5eaHTW/9AzishyPmrnKVsiYiIiCqQdvEGxv77DM7n3+P1EhER6QzHI/3HGWgGLjs7G4WFhYiNjYWJiWo+dPWhHLXtWcqWiIiI6G+rDv6mtozXS0RE9KySkhLk5+fX6Wcs25urtkwQgMS9F+FhIQMAZQx1HUtVOTs7QywW6zoMrWECzcAVFRUBAFq0aKG2jqVsiYiIiDTj9RIREb1Ifn4+4uLi6vQzrnqMAEzM1ZZfuVms9tkrVqyo01iqKiEhQWsF2fQBE2gGzs7ODgCQl5cHDw8PlSqcrZtK8PP/YylbIiIiosp4OkpwPp/XS0REVDlnZ2ckJCTU6WfE7MzF5VuP1ZZ7OlojYWzdfvbLcnZ21nUIWsUEmoHz9vZG06ZNsW7dOhQXF6tU4TRt7gc07Py0fu3/YSlbIiIior/FBHtg3OYzEIS/l/F6iYiIniUWi+t8ptXUflYVjkdT+7WDuzsfKaAPmEAzcKampggICMDOnTtha2uLbmFjcfiWBa7efABxyV043v0ZZs3bokhuwVK2eoYVv4iIag/7VHpZ0nZOWBPpj1WHcnD1ZnGVrpfY3oj0gzbORZ7vpC0vMx6RdjGBZuDKy8uRkZEBd3d3XH3UEKt+LgPw9FkeT8wbo7ixHTrdOY3/LZ8DU1NT3QZLSooKKwqKCitrIv05IBMRVRP7VKopaTunKrcVTe3Nw6KuIiSi52mj7+f4QtpWnfGItI8JNAP3bBXOuH23gOefeSYS4VeTlsjOzoaPj49ugjRSNanE8qIKK9qorlLfKqYQUfVoo9pUVb2oT6xK1araxP5Tv2i7rWpqb7M6P7201pdzh22V9F1dXk9XR2XjjLbHFwWeu0T6SSQIz95ha9zu378PW1tb3Lt3DzY2NroOp1YcOXIEiYmJSElJQadPDuNRabnaNibyUmwMtUevXr10EKHxysnJeelKLFkeIyCvoMKKibwUHX7bVNPQqqS+VUyhumGM/WpNGNP3oyZ9nLZpu09l/6lftN1W9WEMrypjaKvG1K/WFmP6nuj79bSuzndjOHeJDEV1+lTOQDNwz1bhrKyKlEVJEezsPLQdmtGrSSUWfaiwUt8qphBR9Wij2lRt0Xafyv5Tv2i7rerDGF5VbKuk7/T9elpX5zvPXSL9xASagVNU4dy+fTvGv/UPjE85C9U5hQLayP+Et/cIXYVotGpSiYUVVohI32mj2lRtYZ9av2m7rbK9EdUefb+e5vlORM8y0XUAVDOmpqaIiorCmTNncPY/SYgPdoJfcxtYmInQxEQG9+vpmB7RjwUE9Iyiwkp7l0Zo2MAU7V0a4atIf1ZYISJ6CexTSZvY3oj0gzbORZ7vRPQsPgPNSJw4cQLJyckoLCxULnN0dMTIkSPRrVs3HUZGRMbMmPvVl8HvBxFR7WK/qo7fEyKi2sNnoNVD3bp1Q5cuXZCdnY2ioiLY2dnB29ubM8+IiIiIiIiIiGqICTQjYmpqCh8fH12HQURERERERERkVPgMNCIiIiIiIiIiIg2YQCMiIiIiIiIiItLAYBJob731Flq0aAELCws0a9YMw4cPx/Xr13UdFhERERERERERGTmDSaAFBwcjNTUVly9fxvbt25GTk4OwsDBdh0VEREREREREREbOYIoIxMbGKv/fsmVLzJw5E2+//TZKS0thbm6uw8iIiIiIiIiIiMiYGUwC7Vl3795FSkoKAgICNCbPSkpKUFJSonx9//59bYRHRET1BMcZIiKqaxxriIj0g8HcwgkAM2bMgJWVFezt7ZGXl4cffvhB4/aLFi2Cra2t8svFxUVLkRIRUX3AcYaIiOoaxxoiIv0gEgRB0NWHz5s3D/Pnz9e4zalTp9C5c2cAwO3bt3H37l38+eefmD9/PmxtbbFr1y6IRKIK31vRX2tcXFxw79492NjY1N6BEBHVU/fv34etrW297Vc5zhAR1a36Ps4AHGuIiOpSdcYZnd7C+cEHH+C9997TuI2rq6vy/w4ODnBwcICnpye8vb3h4uKCEydOoHv37hW+VywWQywW12bIREREShxniIiornGsISLSDzpNoCkSYi9DMXHu2b/GEBERERERERER1TaDKCKQmZmJzMxM9OzZE3Z2dvj9998RHx8Pd3f3SmefVUSRdOODN4mIaoeiP9Xh0wD0CscZIqLaxXFGHccaIqLaU51xxiASaJaWltixYwfmzp0LmUyGZs2aoV+/ftiyZUu1pjMXFxcDAB+8SURUy4qLi2Fra6vrMHSO4wwRUd3gOPM3jjVERLWvKuOMTosIaJtcLsf169dhbW1daeEBQ6d4qOi1a9f4UFEDxp+jcagPP0dBEFBcXIxXXnkFJiYGVdi5TtSHcUYX6sO5RMaBbbX2cZxRx7Gm9vHcJUPBtlr7qjPOGMQMtNpiYmICZ2dnXYehFTY2NjyhjAB/jsbB2H+OnBHwt/o0zuiCsZ9LZDzYVmsXxxlVHGvqDs9dMhRsq7WrquMM/4xDRERERERERESkARNoREREREREREREGjCBZmTEYjHmzp1breIKpH/4czQO/DkS1Q6eS2Qo2FaJDBPPXTIUbKu6Va+KCBAREREREREREVUXZ6ARERERERERERFpwAQaERERERERERGRBkygERERERERERERacAEGhERERmkqKgovP3227oOgwyQIAgYM2YMGjduDJFIhKysLJ3EkZubq9PPJ6KX4+rqiuXLlytfi0Qi7Ny5U2fxEFVHUFAQpkyZouswDBITaAYqIyMDpqam6NevX6XbfPPNNzA1NcW4ceO0GBnV1I0bNzBx4kS0atUKYrEYLi4uCA0NRXp6uq5Doyqo7Bf6Q4cOQSQS4a+//tJ6TEREpGrv3r1ITk7Grl27UFBQAB8fH12HRFSvFRYWYuzYsWjRogXEYjGcnJwglUpx/PhxAPqfoCooKMAbb7yh6zBIR0QikcavqKgoXYdItcRM1wHQy9mwYQMmTpyIdevWIS8vDy1atKhwm+nTp2P16tVYtmwZGjZsqINIqTpyc3PRo0cPNGrUCEuWLIGfnx9KS0uRlpaGCRMm4Ndff9V1iERERAYvJycHzZo1Q0BAgK5DISIAQ4YMQWlpKTZu3IhWrVrh5s2bSE9Px927d3UdWpU4OTnpOgTSoYKCAuX/t27divj4eFy+fFm5zNLSUmX70tJSmJubay0+qj2cgWaAZDIZUlNTMX78eAwYMADJyclq2+Tm5iIjIwMzZ85EmzZtsG3bNu0HStUWExMDkUiEzMxMhIWFwdPTE+3atcPUqVNx4sQJXYdHRPTSgoKCMHHiREyZMgV2dnZwdHTE119/DZlMhn/84x+wtraGu7s79uzZAwAoLy9HdHQ03NzcYGlpCS8vL6xYsULjZwiCgCVLlqBVq1awtLRE+/btOf6RmqioKEycOBF5eXkQiURwdXV9YdtRzCJOS0tDx44dYWlpiddeew2FhYXYs2cPvL29YWNjg/DwcDx8+FD5vr1796Jnz55o1KgR7O3tMWDAAOTk5GiM79KlS+jfvz8kEgkcHR0xfPhw3L59u86+H0S69tdff+Ho0aNYvHgxgoOD0bJlS3Tt2hWzZs3Cm2++CVdXVwDAoEGDlOcs8DQRPnDgQDg6OkIikaBLly44cOCAcr+zZs1Ct27d1D7Pz88Pc+fOBVDxrWxvv/22yoyhwsJChIaGwtLSEm5ubkhJSVHb57Mz5BS3Zu/YsQPBwcFo2LAh2rdvr5xNp5CRkYHevXvD0tISLi4umDRpEmQyWTW/e6QPnJyclF+2trYQiUTK148fP0ajRo2QmpqKoKAgWFhYYPPmzZg3bx46dOigsp/ly5cr23daWhosLCzU7l6ZNGkSAgMDAQB37txBeHg4nJ2d0bBhQ/j6+uLbb79V2V4mk2HEiBGQSCRo1qwZli5dqhb/5s2b0blzZ1hbW8PJyQnDhg1DYWFhrX1/jAkTaAZo69at8PLygpeXFyIjI5GUlARBEFS22bBhA958803Y2toiMjIS69ev11G0VFV3797F3r17MWHCBFhZWamtb9SokfaDIiKqRRs3boSDgwMyMzMxceJEjB8/HkOHDkVAQADOnj0LqVSK4cOH4+HDh5DL5XB2dkZqaiouXbqE+Ph4/POf/0Rqamql+//oo4+QlJSE1atX4+LFi4iNjUVkZCQOHz6sxaMkfbdixQp8/PHHcHZ2RkFBAU6dOlXltjNv3jx88cUXyMjIwLVr1/DOO+9g+fLl+Oabb7B7927s378fK1euVG4vk8kwdepUnDp1Cunp6TAxMcGgQYMgl8srjK2goACBgYHo0KEDTp8+jb179+LmzZt455136vR7QqRLEokEEokEO3fuRElJidr6U6dOAQCSkpKU5ywAPHjwAP3798eBAwdw7tw5SKVShIaGIi8vDwAQERGBkydPqiStL168iF9++QURERFVji8qKgq5ubn46aefsG3bNqxatapKyYXZs2dj2rRpyMrKgqenJ8LDw1FWVgYA+OWXXyCVSjF48GD8/PPP2Lp1K44ePYoPPvigynGRYZkxYwYmTZqE7OxsSKXSF24fEhKCRo0aYfv27cpl5eXlSE1NVbbfx48fw9/fH7t27cKFCxcwZswYDB8+HCdPnlS+Jy4uDgcPHsT333+Pffv24dChQzhz5ozKZz158gQLFizA+fPnsXPnTvzxxx+87bQyAhmcgIAAYfny5YIgCEJpaang4OAg7N+/X7m+vLxccHFxEXbu3CkIgiDcunVLMDc3F65evaqTeKlqTp48KQAQduzYoetQqAZGjhwpmJqaClZWVipfFhYWAgChqKhI1yES6URgYKDQs2dP5euysjLByspKGD58uHJZQUGBAEA4fvx4hfuIiYkRhgwZonw9cuRIYeDAgYIgCMKDBw8ECwsLISMjQ+U90dHRQnh4eC0eCRmDxMREoWXLloIgVK3tHDx4UAAgHDhwQLl+0aJFAgAhJydHuWzs2LGCVCqt9HMLCwsFAMIvv/wiCIIg/PHHHwIA4dy5c4IgCMKcOXOEvn37qrzn2rVrAgDh8uXLL328RPpu27Ztgp2dnWBhYSEEBAQIs2bNEs6fP69cD0D4/vvvX7iftm3bCitXrlS+9vPzEz7++GPl61mzZgldunRRvg4MDBQmT56sso+BAwcKI0eOFARBEC5fviwAEE6cOKFcn52dLQAQEhMTK4xPcV6vW7dOuf7ixYsCACE7O1sQBEEYPny4MGbMGJXPPXLkiGBiYiI8evTohcdJ+ispKUmwtbVVvla0B8Xv7wpz584V2rdvr7Ls2bFJEARh0qRJwmuvvaZ8nZaWJjRo0EC4e/dupZ/fv39/4cMPPxQEQRCKi4uFBg0aCFu2bFGuv3PnjmBpaanW7p+VmZkpABCKi4s1HGn9xBloBuby5cvIzMzEe++9BwAwMzPDu+++iw0bNii32bdvH2QymfJBlg4ODujbt6/KNqR/hP+bRSgSiXQcCdVUcHAwsrKyVL7WrVun67CIdM7Pz0/5f1NTU9jb28PX11e5zNHREQCUf9lfs2YNOnfujCZNmkAikWDt2rXKmQXPu3TpEh4/fow+ffooZzNIJBJs2rTphbfMUf1WnbbzbBt2dHREw4YN0apVK5Vlz85MycnJwbBhw9CqVSvY2NjAzc0NACptx2fOnMHBgwdV4mjTpo1yX0TGasiQIbh+/Tr+85//QCqV4tChQ+jUqVOFj6pRkMlkmD59Otq2bYtGjRpBIpHg119/VTm/IiIilLdcCoKAb7/9tlqzz7Kzs2FmZobOnTsrl7Vp06ZKd4Y82180a9YMwN/j25kzZ5CcnKxyrkulUsjlcvzxxx9Vjo8Mx7NtqKoiIiJw6NAhXL9+HQCQkpKC/v37w87ODsDTGWkLFy6En58f7O3tIZFIsG/fPuU5kJOTgydPnqB79+7KfTZu3BheXl4qn3Pu3DkMHDgQLVu2hLW1NYKCggBUPlbVZywiYGDWr1+PsrIyNG/eXLlMEASYm5ujqKgIdnZ22LBhA+7evatSNEAul+PcuXNYsGABTE1NdRE6vUDr1q0hEomQnZ1dYRVHMhxWVlbw8PBQWZafn6+jaIj0x/MPzBWJRCrLFH9AkMvlSE1NRWxsLJYuXYru3bvD2toaCQkJKrclPEtxS9zu3btVxkgAEIvFtXkYZGSq03aeb68Vtelnb88MDQ2Fi4sL1q5di1deeQVyuRw+Pj548uRJpbGEhoZi8eLFausUv4ATGSsLCwv06dMHffr0QXx8PN5//33MnTu30lvJ4uLikJaWhs8++wweHh6wtLREWFiYyvk1bNgwzJw5E2fPnsWjR49w7do15UQEADAxMVF7FE5paany/zX5A3dl45vi37Fjx2LSpElq76uoOBwZvucf0fOitgcAXbt2hbu7O7Zs2YLx48fj+++/R1JSknL90qVLkZiYiOXLl8PX1xdWVlaYMmWK8hx4fv8Vkclk6Nu3L/r27YvNmzejSZMmyMvLg1QqrXSsqs+YQDMgZWVl2LRpE5YuXYq+ffuqrBsyZAhSUlIQHh6OH374AVu2bEG7du2U6+VyOXr16oU9e/ZgwIAB2g6dqqBx48aQSqX48ssvMWnSJLVO9q+//uJz0Iio3jhy5AgCAgIQExOjXKZpBk7btm0hFouRl5enfLguUVXUVdu5c+cOsrOz8dVXX6FXr14AgKNHj2p8T6dOnbB9+3a4urrCzIyX6VS/tW3bVvlgfnNzc5SXl6usP3LkCKKiojBo0CAAT5+Jlpubq7KNs7MzevfujZSUFDx69AghISHK2c4A0KRJE5UKiuXl5bhw4QKCg4MBAN7e3igrK8Pp06fRtWtXAE/vCHr+we7V1alTJ1y8eFHtD65UfzRp0gQ3btyAIAjKBGtWVpbadsOGDUNKSgqcnZ1hYmKCN998U7nuyJEjGDhwICIjIwE8/Z3/6tWr8Pb2BgB4eHjA3NwcJ06cUCZmi4qKcOXKFeV49+uvv+L27dv49NNP4eLiAgA4ffp0nR23oePIbEB27dqFoqIiREdHw9bWVmVdWFiYcnaavb09hg4dChMT1Tt0BwwYgPXr1zOBpsdWrVqFgIAAdO3aFR9//DH8/PxQVlaG/fv3Y/Xq1cjOztZ1iEREWuHh4YFNmzYhLS0Nbm5u+Pe//41Tp04pb4F7nrW1NaZNm4bY2FjI5XL07NkT9+/fR0ZGBiQSCUaOHKnlIyBDUVdtx87ODvb29vj666/RrFkz5OXlYebMmRrfM2HCBKxduxbh4eGIi4uDg4MDfvvtN2zZsgVr167lXQRklO7cuYOhQ4di1KhR8PPzg7W1NU6fPo0lS5Zg4MCBAABXV1ekp6ejR48eEIvFsLOzg4eHB3bs2IHQ0FCIRCLMmTOnwgIdERERmDdvHp48eYLExESVda+99hqmTp2K3bt3w93dHYmJiSrJMS8vL/Tr1w+jR4/G119/DTMzM0yZMgWWlpY1OuYZM2agW7dumDBhAkaPHg0rKytkZ2erFSIh4xUUFIRbt25hyZIlCAsLw969e7Fnzx7Y2NiobBcREYH58+dj4cKFCAsLg4WFhXKdh4cHtm/fjoyMDNjZ2WHZsmW4ceOGMoEmkUgQHR2NuLg42Nvbw9HREbNnz1bJE7Ro0QINGjTAypUrMW7cOFy4cAELFizQzjfBAPEZaAZk/fr1CAkJUUueAU9noGVlZSE+Ph6DBg1SS54pttm1axdu3rypjXDpJbi5ueHs2bMIDg7Ghx9+CB8fH/Tp0wfp6elYvXq1rsMjItKacePGYfDgwXj33Xfx6quv4s6dOyqz0SqyYMECxMfHY9GiRfD29oZUKsWPP/5YadKNSKEu2o6JiQm2bNmCM2fOwMfHB7GxsUhISND4nldeeQXHjh1DeXk5pFIpfHx8MHnyZNja2lZ4bUdkDCQSCV599VUkJiaid+/e8PHxwZw5czB69Gh88cUXAJ7eqrZ//364uLigY8eOAIDExETY2dkhICAAoaGhkEql6NSpk9r+hw4dijt37uDhw4dqj0kZNWoURo4ciREjRiAwMBBubm7K2WcKSUlJcHFxQWBgIAYPHowxY8agadOmNTpmPz8/HD58GFevXkWvXr3QsWNHzJkzh7dq1yPe3t5YtWoVvvzyS7Rv3x6ZmZmYNm2a2natW7dGly5d8PPPP6s9v2/OnDno1KkTpFIpgoKC4OTkpNbGExIS0Lt3b7z11lsICQlBz5494e/vr1zfpEkTJCcn47vvvkPbtm3x6aef4rPPPquTYzYGIqEqN8YSERERERERERHVU/xTFhERERERERERkQZMoBEREREREREREWnABBoREREREREREZEGTKARERERERERERFpwAQaERERERERERGRBkygERERERERERERacAEGhERERERERERkQZMoBEREREREREREWnABBoRERERERERVUlGRgZMTU3Rr18/leW5ubkQiUTKL1tbW3Tr1g0//vgjACAoKEhl/fNfrq6uOjgaoqpjAo1Iy548eaLrEIiIiIiIiF7Khg0bMHHiRBw9ehR5eXlq6w8cOICCggKcPHkSXbt2xZAhQ3DhwgXs2LEDBQUFKCgoQGZmpsq2BQUFOHXqlLYPhahamEAjqmNBQUH44IMPMHXqVDg4OKBPnz5YtmwZfH19YWVlBRcXF8TExODBgwcq7zt27BgCAwPRsGFD2NnZQSqVoqioCAAgCAKWLFmCVq1awdLSEu3bt8e2bdt0cXhERKRntm3bBl9fX1haWsLe3h4hISGQyWQAgKSkJHh7e8PCwgJt2rTBqlWrlO8bNWoU/Pz8UFJSAgAoLS2Fv78/IiIidHIcRESkf2QyGVJTUzF+/HgMGDAAycnJatvY29vDyckJbdq0wcKFC1FaWoqDBw+icePGcHJygpOTE5o0aaKy7bPLiPQVE2hEWrBx40aYmZnh2LFj+Oqrr2BiYoLPP/8cFy5cwMaNG/HTTz9h+vTpyu2zsrLw+uuvo127djh+/DiOHj2K0NBQlJeXAwA++ugjJCUlYfXq1bh48SJiY2MRGRmJw4cP6+oQiYhIDxQUFCA8PByjRo1CdnY2Dh06hMGDB0MQBKxduxazZ8/GwoULkZ2djU8++QRz5szBxo0bAQCff/45ZDIZZs6cCQCYM2cObt++rZJkIyKi+m3r1q3w8vKCl5cXIiMjkZSUBEEQKty2tLQUa9euBQCYm5trM0yiOiESKmvtRFQrgoKCcO/ePZw7d67Sbb777juMHz8et2/fBgAMGzYMeXl5OHr0qNq2MpkMDg4O+Omnn9C9e3fl8vfffx8PHz7EN998U/sHQUREBuHs2bPw9/dHbm4uWrZsqbKuRYsWWLx4McLDw5XL/vWvf+G///0vMjIyAADHjx9HYGAgZs6ciUWLFiE9PR29e/fW6jEQEZH+6tGjB9555x1MnjwZZWVlaNasGb799luEhIQgNzcXbm5usLS0hImJCR49egS5XA5XV1ecOXMGjRs3Vu5Hse25c+fQoUMH3R0QUTWY6ToAovqgc+fOKq8PHjyITz75BJcuXcL9+/dRVlaGx48fQyaTwcrKCllZWRg6dGiF+7p06RIeP36MPn36qCx/8uQJOnbsWGfHQERE+q99+/Z4/fXX4evrC6lUir59+yIsLAxlZWW4du0aoqOjMXr0aOX2ZWVlsLW1Vb7u3r07pk2bhgULFmDGjBlMnhERkdLly5eRmZmJHTt2AADMzMzw7rvvYsOGDQgJCVFut3XrVrRp0wZXrlzBlClTsGbNGpXkGZGhYgKNSAusrKyU///zzz/Rv39/jBs3DgsWLEDjxo1x9OhRREdHo7S0FABgaWlZ6b7kcjkAYPfu3WjevLnKOrFYXAfRExGRoTA1NcX+/fuRkZGBffv2YeXKlZg9e7ayAtratWvx6quvqr1HQS6X49ixYzA1NcXVq1e1GjsREem39evXo6ysTOV3EEEQYG5urnxWMwC4uLigdevWaN26NSQSCYYMGYJLly6hadOmugibqNbwGWhEWnb69GmUlZVh6dKl6NatGzw9PXH9+nWVbfz8/JCenl7h+9u2bQuxWIy8vDx4eHiofLm4uGjjEIiISI+JRCL06NED8+fPx7lz59CgQQMcO3YMzZs3x++//642dri5uSnfm5CQgOzsbBw+fBhpaWlISkrS4ZEQEZG+KCsrw6ZNm7B06VJkZWUpv86fP4+WLVsiJSWlwvcFBgbCx8cHCxcu1HLERLWPM9CItMzd3R1lZWVYuXIlQkNDcezYMaxZs0Zlm1mzZsHX1xcxMTEYN24cGjRogIMHD2Lo0KFwcHDAtGnTEBsbC7lcjp49e+L+/fvIyMiARCLByJEjdXRkRESkaydPnkR6ejr69u2Lpk2b4uTJk7h16xa8vb0xb948TJo0CTY2NnjjjTdQUlKC06dPo6ioCFOnTkVWVhbi4+Oxbds29OjRAytWrMDkyZMRGBiIVq1a6frQiIhIh3bt2oWioiJER0er3PoPAGFhYVi/fj0GDBhQ4Xs//PBDDB06FNOnT1e7g4bIkHAGGpGWdejQAcuWLcPixYvh4+ODlJQULFq0SGUbT09P7Nu3D+fPn0fXrl3RvXt3/PDDDzAze5rzXrBgAeLj47Fo0SJ4e3tDKpXixx9/VJlFQERE9Y+NjQ3+97//oX///vD09MRHH32EpUuX4o033sD777+PdevWITk5Gb6+vggMDERycjLc3Nzw+PFjREREICoqCqGhoQCA6OhohISEYPjw4coq0EREVD+tX78eISEhaskzABgyZAiysrJw9+7dCt87YMAAuLq6chYaGTxW4SQiIiIiIiIiItKAM9CIiIiIiIiIiIg0YAKNiIiIiIiIiIhIAybQiIiIiIiIiIiINGACjYiIiIiIiIiISAMm0IiIiIiIiIiIiDRgAo2IiIiIiIiIiEgDJtCIiIiIiIiIiIg0YAKNiIiIiIiIiIhIAybQiIiIiIiIiIiINGACjYiIiIiIiIiISAMm0IiIiIiIiIiIiDRgAo2IiIiIiIiIiEiD/w9xma83H7uD+QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (race_ax, sex_ax, art_ax) = plt.subplots(1,3, sharey=True, figsize = (15, 5))\n",
    "\n",
    "sns.stripplot(data=data,\n",
    "            x = 'race',\n",
    "            y = 'exec_domain_z', ax=race_ax)\n",
    "sns.boxplot(data=data,\n",
    "            x = 'race',\n",
    "            y = 'exec_domain_z', ax=race_ax)\n",
    "\n",
    "sns.stripplot(data=data,\n",
    "            x = 'sex',\n",
    "            y = 'exec_domain_z', ax=sex_ax)\n",
    "sns.boxplot(data=data,\n",
    "            x = 'sex',\n",
    "            y = 'exec_domain_z', ax=sex_ax)\n",
    "\n",
    "sns.stripplot(data=data,\n",
    "            x = 'ART',\n",
    "            y = 'exec_domain_z', ax=art_ax)\n",
    "sns.boxplot(data=data,\n",
    "            x = 'ART',\n",
    "            y = 'exec_domain_z', ax=art_ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6715b3c-a00e-42e2-8633-798881ae7cbb",
   "metadata": {
    "deletable": false,
    "editable": false
   },
   "source": [
    "### Q2: By inspection, which variable has the most between class difference?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b4bacf5-e194-4225-b3c1-3d25c2a830dd",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "|               |    |\n",
    "| --------------|----|\n",
    "| Points        | 5  |\n",
    "| Public Checks | 3  |\n",
    "\n",
    "_Points:_ 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "737a1795-88d3-4225-b0df-e6aee178968a",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [],
   "source": [
    "# Answer: race, sex, ART\n",
    "q2_most_bcd = 'race' # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "788a69b0",
   "metadata": {
    "deletable": false,
    "editable": false
   },
   "outputs": [],
   "source": [
    "grader.check(\"q2_initial_bcd\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27d11168-ead2-4651-b420-a7431b290ee4",
   "metadata": {},
   "source": [
    "## Basic regression"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89603733-b40c-4d31-8ec3-d1933d2e6dd6",
   "metadata": {},
   "source": [
    "We'll start by taking the simplest approach and regress the most correlated value first."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95b2c235-e31d-4198-960c-9759c8cf380a",
   "metadata": {},
   "source": [
    "`pg.linear_regression` works by regressing all columns in the first parameter against the single column in the second.\n",
    "By convention, we usually use the variables `X` and `y`.\n",
    "\n",
    "You'll often see this written as:\n",
    "\n",
    "$\\mathbf{y} = \\mathbf{X} \\boldsymbol{\\beta} + \\boldsymbol{\\epsilon}$\n",
    "\n",
    "In the case of `pg.linear_regression` the $\\boldsymbol{\\epsilon}$ is added by default and we do not need to specify it.\n",
    "\n",
    "You do not have to use the variable names `X` and `y`, in many cases you might have multiple `X`s and `y`s, but for simplicity, I will stick with this simple convention."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "d37176f0-9513-44c9-a293-0256c7f4c08c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>0.711625</td>\n",
       "      <td>0.105822</td>\n",
       "      <td>6.724733</td>\n",
       "      <td>7.994463e-11</td>\n",
       "      <td>0.236815</td>\n",
       "      <td>0.234453</td>\n",
       "      <td>0.503437</td>\n",
       "      <td>0.919812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.035258</td>\n",
       "      <td>0.003522</td>\n",
       "      <td>-10.011320</td>\n",
       "      <td>1.000644e-20</td>\n",
       "      <td>0.236815</td>\n",
       "      <td>0.234453</td>\n",
       "      <td>-0.042186</td>\n",
       "      <td>-0.028329</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names      coef        se          T          pval        r2  \\\n",
       "0          Intercept  0.711625  0.105822   6.724733  7.994463e-11  0.236815   \n",
       "1  YearsSeropositive -0.035258  0.003522 -10.011320  1.000644e-20  0.236815   \n",
       "\n",
       "     adj_r2  CI[2.5%]  CI[97.5%]  \n",
       "0  0.234453  0.503437   0.919812  \n",
       "1  0.234453 -0.042186  -0.028329  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data['YearsSeropositive'] # Our independent variables\n",
    "y = data['exec_domain_z']     # Our dependent variable\n",
    "res = pg.linear_regression(X, y)\n",
    "res"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "308f2c65-40b8-4e26-93a9-2b2ac44e495f",
   "metadata": {},
   "source": [
    "This has fit the equation:\n",
    "\n",
    "`PDZ = -0.035*YS + 0.712`\n",
    "\n",
    "It tells us that the likelihood of this slope being zero is 1.0E-20 and that years-seropositive explains ~23.6% of variation in EDZ that we observe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "f97f1fce-b27c-4371-bc5e-97378e170ff5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fafb72201f0>]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.regplot(data = data,\n",
    "                 x = 'YearsSeropositive',\n",
    "                 y = 'exec_domain_z')\n",
    "\n",
    "# Pick \"years seropositive\" from 0 to 70\n",
    "x = np.arange(0, 70)\n",
    "\n",
    "# Use the coefficients from above in a linear equation\n",
    "y = res.loc[1, 'coef']*x + res.loc[0, 'coef']\n",
    "\n",
    "ax.plot(x, y, color = 'r')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b9d1f9b-16b9-4f95-ae29-00d964a2eb3c",
   "metadata": {},
   "source": [
    "## Residuals"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9909e11-b673-4be1-9787-e4f815f04ab7",
   "metadata": {},
   "source": [
    "_Residuals_ are the difference between the observed value and the predicted value.\n",
    "In the case of a simple linear regression, this is the y-distance between each point and the best-fit line.\n",
    "Examining these is an import step in assessing the fit for any biases.\n",
    "You can think of the residual as what is \"left over\" after the regression.\n",
    "\n",
    "We could calculate these ourselves from the regression coefficients, but, `pingouin` conviently provides them for us.\n",
    "The result `DataFrame` from `pg.linear_regression` has a special attribute `.residuals_` which stores the difference between the prediction and reality for each point in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "aff2050d-1d24-4b23-834a-dd8e9add1aa0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.34672285  1.15826787 -0.29430717 -1.06544462  1.08198035]\n"
     ]
    }
   ],
   "source": [
    "print(res.residuals_[:5])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2662e02-ff9b-4398-ace9-d4f05d29e098",
   "metadata": {},
   "source": [
    "In order to test the **Homoscedasticity** we want to ensure that these residuals are _not correlated with the depenendant variable_.\n",
    "\n",
    "In our case, this means that the model is equally good predicting the EDZ of people recently infected with HIV and those who have been living with HIV for a long time.\n",
    "\n",
    "To do this, we plot the residuals vs each independent variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2eec2b7c-2bae-4b79-a740-f534751b66e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='YearsSeropositive'>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(x=data['YearsSeropositive'],  y=res.residuals_)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddc1570e-155a-4c57-ac8d-e41eb6895574",
   "metadata": {},
   "source": [
    "This is an ideal residual plot.\n",
    "It should look like a random \"stary-night sky\" centered around 0.\n",
    "This implies that the model is not better or worse for any given X value."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d4a62b5-c418-4222-9c87-90ecf7804f26",
   "metadata": {},
   "source": [
    "Let's also test our assumption about a normal distribution of errors of the residuals."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca391103-3c84-4fd6-9b7f-896577811ed5",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": []
   },
   "source": [
    "### Q3: Are the residuals normally distributed?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41d6da6d-1e4c-496e-a059-85b262326bc9",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "|               |    |\n",
    "| --------------|----|\n",
    "| Points        | 5  |\n",
    "| Public Checks | 5  |\n",
    "\n",
    "_Points:_ 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0caa835c-e80d-4ec1-ba53-de99147c41d5",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a Q-Q plot of the residuals\n",
    "\n",
    "q3_plot = pg.qqplot(res.residuals_)  # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "753e8d3b-8d25-4ac7-81d7-8f606d9dec09",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [],
   "source": [
    "# Use the Jarque-Bera normal test for large sample sizes\n",
    "\n",
    "q3_norm_res = pg.normality(res.residuals_, method='jarque_bera')  # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "5afc057b-0cf0-4df7-8d5e-734980f2fb47",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [],
   "source": [
    "# Are the residuals normally distributed? 'yes' or 'no'\n",
    "\n",
    "q3_is_norm = 'yes' # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f297104c",
   "metadata": {
    "deletable": false,
    "editable": false
   },
   "outputs": [],
   "source": [
    "grader.check(\"q3_resid_normality\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01b59934-9f51-429d-a65e-ebf77655a3dc",
   "metadata": {},
   "source": [
    "You don't need to do this test at every stage, but it is a good test to do before you are _done_."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17cd99fc-7bc7-4f43-9872-50ddc5fc4a9d",
   "metadata": {},
   "source": [
    "## Multiple Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0045aea-276f-4dd8-bfd2-cf9129a2cb15",
   "metadata": {},
   "source": [
    "Regression is not limited to a single independent variable, you can add as many as you'd like.\n",
    "\n",
    "In our case, there are two others that we should consider: `age` and `education`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "2c9e5a55-d612-4af6-a1b2-113e9ae5f825",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>0.977449</td>\n",
       "      <td>0.404718</td>\n",
       "      <td>2.415135</td>\n",
       "      <td>1.628781e-02</td>\n",
       "      <td>0.318207</td>\n",
       "      <td>0.311835</td>\n",
       "      <td>0.181214</td>\n",
       "      <td>1.773685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.037462</td>\n",
       "      <td>0.003390</td>\n",
       "      <td>-11.049854</td>\n",
       "      <td>2.853764e-24</td>\n",
       "      <td>0.318207</td>\n",
       "      <td>0.311835</td>\n",
       "      <td>-0.044132</td>\n",
       "      <td>-0.030792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>-0.102647</td>\n",
       "      <td>0.020406</td>\n",
       "      <td>-5.030176</td>\n",
       "      <td>8.170366e-07</td>\n",
       "      <td>0.318207</td>\n",
       "      <td>0.311835</td>\n",
       "      <td>-0.142794</td>\n",
       "      <td>-0.062500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>0.019297</td>\n",
       "      <td>0.005546</td>\n",
       "      <td>3.479295</td>\n",
       "      <td>5.721793e-04</td>\n",
       "      <td>0.318207</td>\n",
       "      <td>0.311835</td>\n",
       "      <td>0.008385</td>\n",
       "      <td>0.030209</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names      coef        se          T          pval        r2  \\\n",
       "0          Intercept  0.977449  0.404718   2.415135  1.628781e-02  0.318207   \n",
       "1  YearsSeropositive -0.037462  0.003390 -11.049854  2.853764e-24  0.318207   \n",
       "2          education -0.102647  0.020406  -5.030176  8.170366e-07  0.318207   \n",
       "3                age  0.019297  0.005546   3.479295  5.721793e-04  0.318207   \n",
       "\n",
       "     adj_r2  CI[2.5%]  CI[97.5%]  \n",
       "0  0.311835  0.181214   1.773685  \n",
       "1  0.311835 -0.044132  -0.030792  \n",
       "2  0.311835 -0.142794  -0.062500  \n",
       "3  0.311835  0.008385   0.030209  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data[['YearsSeropositive', 'education', 'age']]\n",
    "y = data['exec_domain_z']\n",
    "res = pg.linear_regression(X, y)\n",
    "res"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3653f050-b236-46ff-8b0d-4db6935c6880",
   "metadata": {},
   "source": [
    "Now, it has fit the equation:\n",
    "\n",
    "`EDZ = -0.037*YS - 0.103*edu + 0.019*age + 0.977`\n",
    "\n",
    "The education is significant at p=8.17E-7.\n",
    "Be caution when comparing coefficients, we might be tempted to compare -0.0422 and -0.0506 and say that education has a more negative effect than YS ...\n",
    "But, remember that education ranges from 0-12 and YS ranges from 0-60, these are not on the same scale and are not directly comparable.\n",
    "We'll talk about how to compare relative importance later."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60eb2693-5c50-4784-889d-ac28a1faba2b",
   "metadata": {},
   "source": [
    "As before, we should check the residuals of the model against _each_ independent variable in the regression to check for homoscedasticity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "d131c037-88eb-491d-a707-8526b6d2c516",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='age'>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ys_ax, edu_ax, age_ax) = plt.subplots(1,3, sharey=True, figsize = (15, 5))\n",
    "\n",
    "sns.scatterplot(x=data['YearsSeropositive'],  y=res.residuals_, ax=ys_ax)\n",
    "sns.scatterplot(x=data['education'],  y=res.residuals_, ax=edu_ax)\n",
    "sns.scatterplot(x=data['age'],  y=res.residuals_, ax=age_ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e162e5c1-107e-4d83-a074-8d9812b67688",
   "metadata": {},
   "source": [
    "Three more stary night skies. Perfect."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dc72fe5-e59a-434b-acba-3ceacd58ecfe",
   "metadata": {},
   "source": [
    "Remember, the residual is the difference between the prediction of the model and reality.\n",
    "Therefore, we can also use the residual plots to see how well the regression is handling other variables we have not included in the model.\n",
    "If the model has properly accounted for something, the residual plot should stay centered around 0.\n",
    "\n",
    "This can be done for categorical or continious variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "15d2e733-b303-4aff-8451-147f222f5cd7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='ART'>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABN0AAAHACAYAAACWIh5pAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA8yUlEQVR4nO3df7zX8/0//tvpdzp1pFRMlIUy5Uek8iO2lFFr+f0rTBj5bRhvy5j31saizYYtFNP8GMbM1o8Z3kgoNUNre/vMam8lP1Ic9PP1/cO3M0dF8qzTqev1cnldnMfz+Xg+X/fnOc/Xeeh2Hs/ns6xUKpUCAAAAABSmTk0XAAAAAAAbGqEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwerVdAHru2XLluXVV19N06ZNU1ZWVtPlANR6pVIp77zzTrbccsvUqeNvP8YZgGIZZ1ZkrAEo1uqONUK3T/Hqq6+mbdu2NV0GwAZn1qxZ2WqrrWq6jBpnnAFYO4wz/2GsAVg7Pm2sEbp9iqZNmyb58BvZrFmzGq4GoPZbsGBB2rZtW/X7dWNnnAEolnFmRcYagGKt7lgjdPsUy6dfN2vWzAAFUCCXt3zIOAOwdhhn/sNYA7B2fNpY4yYHAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwerVdAHUrFKplMrKyqp2kyZNUlZWVoMVAQAAANR+QreNXGVlZQYMGFDVfuCBB1JeXl6DFQEAAADUfi4vBQAAAICCCd0AAAAAoGAuLwUAAADYwLiHe80TugEAAABsYNzDvea5vBQAAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAggndAAAAAKBgQjcAAAAAKJjQDQAAAAAKJnQDAAAAgIIJ3QAAAACgYEI3AAAAACiY0A0AAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAggndAAAAAKBgQjcAAAAAKJjQDQAAAAAKJnQDAAAAgIIJ3QAAAACgYEI3AAAAACiY0A0AAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAApWr6YL2NB0vfC2mi7hMylbsigVH2nvN/TOlOo1qLF6PqspVx9f0yUAAAAArMBMNwAAAAAomNANAAAAAAomdAMAAACAgtW60O36669P+/bt06hRo3Tt2jWPP/74Kvved999OeCAA7L55punWbNm6dGjR8aNG7cOqwUAAABgY1SrQre77ror5557bi699NJMnTo1++yzT7761a9m5syZK+3/P//zPznggAPyhz/8IVOmTMn++++f/v37Z+rUqeu4cgAAAAA2JrUqdLvmmmsyePDgnHzyyenUqVNGjBiRtm3b5oYbblhp/xEjRuSiiy7KHnvske222y4/+MEPst122+XBBx9cx5UDAAAAsDGpV9MFrK5FixZlypQpufjii6st79OnTyZOnLha+1i2bFneeeedbLbZZqvss3DhwixcuLCqvWDBgjUrGABWwjgDwNpmrAFYP9SamW5vvPFGli5dmtatW1db3rp168yZM2e19jF8+PBUVlbmiCOOWGWfYcOGpaKiourVtm3bz1U3AHyUcQaAtc1YA7B+qDWh23JlZWXV2qVSaYVlK3PHHXfk8ssvz1133ZVWrVqtst8ll1yS+fPnV71mzZr1uWsGgOWMMwCsbcYagPVDrbm8tGXLlqlbt+4Ks9rmzp27wuy3j7vrrrsyePDg/OY3v0nv3r0/sW/Dhg3TsGHDz10vAKyMcQaAtc1YA7B+qDUz3Ro0aJCuXbtmwoQJ1ZZPmDAhPXv2XOV2d9xxR0488cT8+te/zsEHH7y2ywQAAACA2jPTLUnOP//8DBo0KLvvvnt69OiRX/7yl5k5c2ZOO+20JB9Oo/6///u/3HbbbUk+DNyOP/74/OQnP0n37t2rZsk1btw4FRUVNXYcAAAAAGzYalXoduSRR+bNN9/M9773vcyePTs77bRT/vCHP2SbbbZJksyePTszZ86s6v+LX/wiS5YsyRlnnJEzzjijavkJJ5yQ0aNHr+vyAQAAANhI1KrQLUmGDBmSIUOGrHTdx4O0Rx99dO0XBAAAAAAfU2vu6QYAAAAAtYXQDQAAAAAKJnQDAAAAgILVunu6AQAAAKxK1wtvq+kS1gtlSxal4iPt/YbemVK9BjVWz/piytXHr7P3MtMNAAAAAAomdAMAAACAggndAAAAAKBgQjcAAAAAKJjQDQAAAAAKJnQDAAAAgILVq+kCAABgY1QqlVJZWVnVbtKkScrKymqwIgCgSEI3AACoAZWVlRkwYEBV+4EHHkh5eXkNVgQAFMnlpQAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUrF5NFwAAUKRSqZTKysqqdpMmTVJWVlaDFQEAsDESugEAG5TKysoMGDCgqv3AAw+kvLy8BisCAGBjJHTbyJXq1s/8LkdXawMAAADw+QjdNnZlZSnVa1DTVQAAAABsUDxIAQAAAAAKJnQDAAAAgIIJ3QAAAACgYEI3AAAAACiY0A0AAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAggndAAAAAKBgQjcAAAAAKJjQDQAAAAAKJnQDAAAAgIIJ3QAAAACgYEI3AAAAACiY0A0AAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAgtWr6QIAAAAAKFapbv3M73J0tTbrltANAAAAYENTVpZSvQY1XcVGzeWlAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBal3odv3116d9+/Zp1KhRunbtmscff/wT+z/22GPp2rVrGjVqlG233TY33njjOqoU4LMplUp59913q16lUqmmSwIAAGAN1arQ7a677sq5556bSy+9NFOnTs0+++yTr371q5k5c+ZK+//zn//MQQcdlH322SdTp07Nf/3Xf+Xss8/Ovffeu44rB/h0lZWVGTBgQNWrsrKypksCAABgDdWq0O2aa67J4MGDc/LJJ6dTp04ZMWJE2rZtmxtuuGGl/W+88cZsvfXWGTFiRDp16pSTTz45J510Un784x+v48oBAAAA2JjUmtBt0aJFmTJlSvr06VNteZ8+fTJx4sSVbvPUU0+t0L9v376ZPHlyFi9evNZqBQAAAGDjVq+mC1hdb7zxRpYuXZrWrVtXW966devMmTNnpdvMmTNnpf2XLFmSN954I1tsscUK2yxcuDALFy6sai9YsKCA6gHgQ2trnOl64W2F7GdDULZkUSo+0t5v6J0p1WtQY/WsL6ZcfXxNlwCsI/5NA7B+qDWh23JlZWXV2qVSaYVln9Z/ZcuXGzZsWK644oo1rs//0G4Y/ON13Vpbn5va9nOs7UGB33+r5/OOM6vi+/8f7777bgYMuKOq/eiVR6W8vLwGK+Kjatvv5rWptv/eX1v8Pvv81tZY4/PLJ1mfPrvrUy1s3GrN5aUtW7ZM3bp1V5jVNnfu3BVmsy3Xpk2blfavV69eWrRosdJtLrnkksyfP7/qNWvWrGIOAABinAFg7TPWAKwfas1MtwYNGqRr166ZMGFCBg4cWLV8woQJGTBgwEq36dGjRx588MFqy8aPH5/dd9899evXX+k2DRs2TMOGDYsrHAA+wjjDxs7sg/8wK5O1xVgDsH6oNTPdkuT888/PTTfdlFtuuSXTp0/Peeedl5kzZ+a0005L8uFfdI4//j//I3faaaflX//6V84///xMnz49t9xyS26++eZccMEFNXUIAAAAAGwEas1MtyQ58sgj8+abb+Z73/teZs+enZ122il/+MMfss022yRJZs+enZkzZ1b1b9++ff7whz/kvPPOy89//vNsueWW+elPf5pDDz20pg4BAAAAgI1ArQrdkmTIkCEZMmTISteNHj16hWW9evXKc889t5arAgAAAID/qFWXlwIAAABAbSB0AwAAAICCCd0AAAAAoGBCNwAAAAAoWK17kAIAAADUmFIpZUsX/6dZt35SVlaDBQHrK6EbAAAArKaypYtT8fwdVe35XY5OqV6DGqwIWF+5vBQAAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAggndAAAAAKBgQjcAAAAAKJjQDQAAAAAKJnQDAAAAgIIJ3QAAAACgYEI3AAAAACiY0A0AAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAggndAAAAAKBgQjcAAAAAKJjQDQAAAAAKJnQDAAAAgIIJ3QAAAACgYEI3AAAAACiY0A0AAAAAClavpgsAAChSkyZN8sADD1RrAwDAuiZ0AwA2KGVlZSkvL6/pMgAA2Mi5vBQAAAAACiZ0AwAAAICCCd0AAAAAoGBCNwAAAAAomNANAAAAAAomdAMAAACAgtVb3Y677rprysrKVqvvc889t8YFAQAAAEBtt9qh29e//vW1WAYAAAAAbDhWO3T77ne/uzbrAAAAAIANhnu6AQAAAEDBVnum20ctXbo01157be6+++7MnDkzixYtqrb+rbfeKqQ4AAAAAKiN1mim2xVXXJFrrrkmRxxxRObPn5/zzz8/hxxySOrUqZPLL7+84BIBAAAAoHZZo5luY8aMyciRI3PwwQfniiuuyNFHH50vfvGL6dKlSyZNmpSzzz676DoBAGCD0qRJkzzwwAPV2rA+m3L18TVdwnrh3XffzYABd1S1H73yqJSXl9dgRcD6ao1mus2ZMyedO3dOkpSXl2f+/PlJkn79+uWhhx4qrjoAANhAlZWVpby8vOpVVlZW0yUBAAVao9Btq622yuzZs5MkHTp0yPjx45Mkzz77bBo2bFhcdQAAAABQC61R6DZw4MA8/PDDSZJzzjknQ4cOzXbbbZfjjz8+J510UqEFAgAAAEBts0b3dPvhD39Y9fVhhx2WrbbaKhMnTkyHDh3yta99rbDiAAAAAKA2WqPQ7eO6d++e7t27F7ErAAAAAKj11ih0u+222z5x/fHHe6oNAAAAABuvNQrdzjnnnGrtxYsX57333kuDBg2yySabCN0AAAAA2Kit0YMU5s2bV+317rvvZsaMGdl7771zxx13FF0jAAAAANQqaxS6rcx2222XH/7whyvMggMAAACAjU1hoVuS1K1bN6+++mqRu6wyb968DBo0KBUVFamoqMigQYPy9ttvr7L/4sWL8+1vfzudO3dOkyZNsuWWW+b4449fa/UBAAAAwHJrdE+33/3ud9XapVIps2fPzs9+9rPstddehRT2ccccc0z+/e9/Z+zYsUmSU089NYMGDcqDDz640v7vvfdennvuuQwdOjQ777xz5s2bl3PPPTdf+9rXMnny5LVSIwAAAAAkaxi6ff3rX6/WLisry+abb54vf/nLGT58eBF1VTN9+vSMHTs2kyZNyp577pkkGTlyZHr06JEZM2Zkhx12WGGbioqKTJgwodqy6667Lt26dcvMmTOz9dZbF14nAAAAACRrGLotW7as6Do+0VNPPZWKioqqwC1JunfvnoqKikycOHGlodvKzJ8/P2VlZdl0001X2WfhwoVZuHBhVXvBggVrXDcAfJxxBoC1zVgDsH4o9J5ua8ucOXPSqlWrFZa3atUqc+bMWa19fPDBB7n44otzzDHHpFmzZqvsN2zYsKr7xlVUVKRt27ZrXDcAfJxxBoC1zVgDsH5Y7Zlu559//mrv9JprrlmtfpdffnmuuOKKT+zz7LPPJvnwEtaPK5VKK13+cYsXL85RRx2VZcuW5frrr//Evpdcckm1Y12wYIFBCoDCGGcAWNuMNQDrh9UO3aZOnVqtPWXKlCxdurTq0s6///3vqVu3brp27brab37mmWfmqKOO+sQ+7dq1y/PPP5/XXntthXWvv/56Wrdu/YnbL168OEcccUT++c9/5s9//vMnznJLkoYNG6Zhw4afXjwArAHjDABrm7EGYP2w2qHbI488UvX1Nddck6ZNm+bWW29N8+bNkyTz5s3LN77xjeyzzz6r/eYtW7ZMy5YtP7Vfjx49Mn/+/DzzzDPp1q1bkuTpp5/O/Pnz07Nnz1Vutzxw+8c//pFHHnkkLVq0WO3aAAAAAGBNrdE93YYPH55hw4ZVBW5J0rx58/z3f//3Wnl6aadOnXLggQfmlFNOyaRJkzJp0qSccsop6devX7WHKHTs2DG//e1vkyRLlizJYYcdlsmTJ2fMmDFZunRp5syZkzlz5mTRokWF1wgAAAAAy61R6LZgwYKVXu45d+7cvPPOO5+7qJUZM2ZMOnfunD59+qRPnz7p0qVLfvWrX1XrM2PGjMyfPz9J8u9//zu/+93v8u9//zu77LJLtthii6rXxIkT10qNAAAAAJB8hstLP2rgwIH5xje+keHDh6d79+5JkkmTJuXCCy/MIYccUmiBy2222Wa5/fbbP7FPqVSq+rpdu3bV2gAAAACwrqxR6HbjjTfmggsuyHHHHZfFixd/uKN69TJ48OBcffXVhRYIAAAAALXNGoVum2yySa6//vpcffXVefnll1MqldKhQ4c0adKk6PoAAAAAoNZZo9BtuSZNmqRLly5F1QIAAAAAG4TVDt0OOeSQjB49Os2aNfvU+7bdd999n7swAAAAAKitVjt0q6ioSFlZWdXXAAAAAMDKrXboNmrUqJV+DQAAAABUV2dNNnr//ffz3nvvVbX/9a9/ZcSIERk/fnxhhQEAAABAbbVGoduAAQNy2223JUnefvvtdOvWLcOHD8+AAQNyww03FFogAAAAANQ2a/T00ueeey7XXnttkuSee+5JmzZtMnXq1Nx777257LLLcvrppxdaJMCamHL18TVdwmfy7rvvZsCAO6raj155VMrLy2uwIgAAANbUGs10e++999K0adMkyfjx43PIIYekTp066d69e/71r38VWiAAAAAA1DZrFLp16NAh999/f2bNmpVx48alT58+SZK5c+emWbNmhRYIAAAAALXNGoVul112WS644IK0a9cu3bp1S48ePZJ8OOtt1113LbRAAAAAAKht1uiebocddlj23nvvzJ49OzvvvHPV8q985SsZOHBgYcUBAAAAQG20RjPdkqRNmzZp2rRpJkyYkPfffz9Jsscee6Rjx46FFQcAAAAAtdEahW5vvvlmvvKVr2T77bfPQQcdlNmzZydJTj755HzrW98qtEAAAAAAqG3WKHQ777zzUr9+/cycOTObbLJJ1fIjjzwyY8eOLaw4AAAAAKiN1uiebuPHj8+4ceOy1VZbVVu+3Xbb5V//+lchhQEAAABAbbVGM90qKyurzXBb7o033kjDhg0/d1EAAAAAUJutUei277775rbbbqtql5WVZdmyZbn66quz//77F1YcAAAAANRGa3R56Y9//OP06tUrkydPzqJFi3LRRRflxRdfzFtvvZUnn3yy6BoBAABgvdCkSZM88MAD1doAK/OZQ7fFixdnyJAh+d3vfpc//vGPqVu3biorK3PIIYfkjDPOyBZbbLE26gQAAIAaV1ZWlvLy8pouA6gFPnPoVr9+/bzwwgtp0aJFrrjiirVREwAAAADUamt0T7fjjz8+N998c9G1AAAAAMAGYY3u6bZo0aLcdNNNmTBhQnbfffcVrmG/5pprCikOAAAAAGqjNQrdXnjhhey2225Jkr///e/V1pWVlX3+qgAAAACgFluj0O2RRx4pug4AAAAA2GCs0T3dAAAAAIBVE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABSs1oRu8+bNy6BBg1JRUZGKiooMGjQob7/99mpv/81vfjNlZWUZMWLEWqsRAAAAAJJaFLodc8wxmTZtWsaOHZuxY8dm2rRpGTRo0Gpte//99+fpp5/OlltuuZarBAAAAICkXk0XsDqmT5+esWPHZtKkSdlzzz2TJCNHjkyPHj0yY8aM7LDDDqvc9v/+7/9y5plnZty4cTn44IPXVckAAAAAbMRqRej21FNPpaKioipwS5Lu3bunoqIiEydOXGXotmzZsgwaNCgXXnhhvvSlL63Wey1cuDALFy6sai9YsODzFQ/rQqmUsqWL/9OsWz8pK6vBgoBVMc4AsLYZawDWD7Xi8tI5c+akVatWKyxv1apV5syZs8rtfvSjH6VevXo5++yzV/u9hg0bVnXfuIqKirRt23aNaoZ1qWzp4lQ8f0fV66MBHLB+Mc4AsLYZawDWDzUaul1++eUpKyv7xNfkyZOTJGUrmbVTKpVWujxJpkyZkp/85CcZPXr0KvuszCWXXJL58+dXvWbNmrVmBwcAK2GcAWBtM9YArB9q9PLSM888M0cdddQn9mnXrl2ef/75vPbaayuse/3119O6deuVbvf4449n7ty52XrrrauWLV26NN/61rcyYsSIvPLKKyvdrmHDhmnYsOHqHwQAfAbGGQDWNmMNwPqhRkO3li1bpmXLlp/ar0ePHpk/f36eeeaZdOvWLUny9NNPZ/78+enZs+dKtxk0aFB69+5dbVnfvn0zaNCgfOMb3/j8xQMAAADAKtSKByl06tQpBx54YE455ZT84he/SJKceuqp6devX7WHKHTs2DHDhg3LwIED06JFi7Ro0aLafurXr582bdp84tNOAQAAAODzqhUPUkiSMWPGpHPnzunTp0/69OmTLl265Fe/+lW1PjNmzMj8+fNrqEIAAAAA+FCtmOmWJJtttlluv/32T+xTKpU+cf2q7uMGAAAAAEWqNTPdAAAAAKC2ELoBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMHq1XQBAHyoSZMmeeCBB6q1AQAAqJ2EbgDribKyspSXl9d0GQAAABTA5aUAAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQsHo1XQCsj6ZcfXxNl/CZvPvuuxkw4I6q9qNXHpXy8vIarAgAAAA2bma6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwWpN6DZv3rwMGjQoFRUVqaioyKBBg/L2229/6nbTp0/P1772tVRUVKRp06bp3r17Zs6cufYLBgAAAGCjVWtCt2OOOSbTpk3L2LFjM3bs2EybNi2DBg36xG1efvnl7L333unYsWMeffTR/OUvf8nQoUPTqFGjdVQ1AAAAABujejVdwOqYPn16xo4dm0mTJmXPPfdMkowcOTI9evTIjBkzssMOO6x0u0svvTQHHXRQrrrqqqpl22677TqpGQAAAICNV62Y6fbUU0+loqKiKnBLku7du6eioiITJ05c6TbLli3LQw89lO233z59+/ZNq1atsueee+b+++//xPdauHBhFixYUO0FAEUxzgCwthlrANYPtSJ0mzNnTlq1arXC8latWmXOnDkr3Wbu3Ll5991388Mf/jAHHnhgxo8fn4EDB+aQQw7JY489tsr3GjZsWNV94yoqKtK2bdvCjgMAjDMArG3GGoD1Q42GbpdffnnKyso+8TV58uQkSVlZ2Qrbl0qllS5PPpzpliQDBgzIeeedl1122SUXX3xx+vXrlxtvvHGVNV1yySWZP39+1WvWrFkFHCkAfMg4A8DaZqwBWD/U6D3dzjzzzBx11FGf2Kddu3Z5/vnn89prr62w7vXXX0/r1q1Xul3Lli1Tr1697LjjjtWWd+rUKU888cQq369hw4Zp2LDhalQPAJ+dcQaAtc1YA7B+qNHQrWXLlmnZsuWn9uvRo0fmz5+fZ555Jt26dUuSPP3005k/f3569uy50m0aNGiQPfbYIzNmzKi2/O9//3u22Wabz188AAAAAKxCrbinW6dOnXLggQfmlFNOyaRJkzJp0qSccsop6devX7Unl3bs2DG//e1vq9oXXnhh7rrrrowcOTL/+7//m5/97Gd58MEHM2TIkJo4DAAAAAA2ErUidEuSMWPGpHPnzunTp0/69OmTLl265Fe/+lW1PjNmzMj8+fOr2gMHDsyNN96Yq666Kp07d85NN92Ue++9N3vvvfe6Lh8AAACAjUiNXl76WWy22Wa5/fbbP7FPqVRaYdlJJ52Uk046aW2VBQAAAAArqDUz3QAAAACgthC6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwerVdAHA59ekSZM88MAD1doAAABAzRG6wQagrKws5eXlNV0GAAAA8P9zeSkAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAUDChGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFKxeTRewviuVSkmSBQsW1HAlABuG5b9Pl/9+3dgZZwCKZZxZkbEGoFirO9YI3T7FO++8kyRp27ZtDVcCsGF55513UlFRUdNl1DjjDMDaYZz5D2MNwNrxaWNNWcmfgD7RsmXL8uqrr6Zp06YpKyur6XLWigULFqRt27aZNWtWmjVrVtPlsIb8HDcMG8PPsVQq5Z133smWW26ZOnXc5WBjGGdqwsbwWWLD4FwtnnFmRcaa4vnsUls4V9eO1R1rzHT7FHXq1MlWW21V02WsE82aNfMh3AD4OW4YNvSfo5kH/7ExjTM1YUP/LLHhcK4WyzhTnbFm7fHZpbZwrhZvdcYaf/oBAAAAgIIJ3QAAAACgYEI30rBhw3z3u99Nw4YNa7oUPgc/xw2DnyMUw2eJ2sK5CrWTzy61hXO1ZnmQAgAAAAAUzEw3AAAAACiY0A0AAAAACiZ0AwAAAICCCd0AgI3GiSeemK9//es1XQa1UKlUyqmnnprNNtssZWVlmTZtWo3U8corr9To+wNrpl27dhkxYkRVu6ysLPfff3+N1QOfxX777Zdzzz23psuolYRuG5GJEyembt26OfDAA1fZ59e//nXq1q2b0047bR1Wxuc1Z86cnHXWWdl2223TsGHDtG3bNv3798/DDz9c06WxGlYVAjz66KMpKyvL22+/vc5rAqC6sWPHZvTo0fn973+f2bNnZ6eddqrpkmCjNnfu3Hzzm9/M1ltvnYYNG6ZNmzbp27dvnnrqqSTrf6g1e/bsfPWrX63pMqghZWVln/g68cQTa7pEClKvpgtg3bnlllty1lln5aabbsrMmTOz9dZbr7TPRRddlBtuuCHXXHNNNtlkkxqolM/ilVdeyV577ZVNN900V111Vbp06ZLFixdn3LhxOeOMM/K3v/2tpksEgFrv5ZdfzhZbbJGePXvWdClAkkMPPTSLFy/Orbfemm233TavvfZaHn744bz11ls1XdpqadOmTU2XQA2aPXt21dd33XVXLrvsssyYMaNqWePGjav1X7x4cerXr7/O6qM4ZrptJCorK3P33Xfn9NNPT79+/TJ69OgV+rzyyiuZOHFiLr744nTs2DH33HPPui+Uz2zIkCEpKyvLM888k8MOOyzbb799vvSlL+X888/PpEmTaro8gDW233775ayzzsq5556b5s2bp3Xr1vnlL3+ZysrKfOMb30jTpk3zxS9+MX/84x+TJEuXLs3gwYPTvn37NG7cODvssEN+8pOffOJ7lEqlXHXVVdl2223TuHHj7LzzzsY/VnDiiSfmrLPOysyZM1NWVpZ27dp96rmzfLbyuHHjsuuuu6Zx48b58pe/nLlz5+aPf/xjOnXqlGbNmuXoo4/Oe++9V7Xd2LFjs/fee2fTTTdNixYt0q9fv7z88sufWN9LL72Ugw46KOXl5WndunUGDRqUN954Y619P6Cmvf3223niiSfyox/9KPvvv3+22WabdOvWLZdcckkOPvjgtGvXLkkycODAqs9s8mF4PmDAgLRu3Trl5eXZY4898qc//alqv5dcckm6d+++wvt16dIl3/3ud5Os/DK7r3/969VmJs2dOzf9+/dP48aN0759+4wZM2aFfX50Jt7yy8bvu+++7L///tlkk02y8847V83aW27ixInZd99907hx47Rt2zZnn312KisrP+N3j/VBmzZtql4VFRUpKyuran/wwQfZdNNNc/fdd2e//fZLo0aNcvvtt+fyyy/PLrvsUm0/I0aMqDq/x40bl0aNGq1wlczZZ5+dXr16JUnefPPNHH300dlqq62yySabpHPnzrnjjjuq9a+srMzxxx+f8vLybLHFFhk+fPgK9d9+++3Zfffd07Rp07Rp0ybHHHNM5s6dW9j3Z0MidNtI3HXXXdlhhx2yww475LjjjsuoUaNSKpWq9bnlllty8MEHp6KiIscdd1xuvvnmGqqW1fXWW29l7NixOeOMM9KkSZMV1m+66abrviiAAt16661p2bJlnnnmmZx11lk5/fTTc/jhh6dnz5557rnn0rdv3wwaNCjvvfdeli1blq222ip33313XnrppVx22WX5r//6r9x9992r3P93vvOdjBo1KjfccENefPHFnHfeeTnuuOPy2GOPrcOjZH33k5/8JN/73vey1VZbZfbs2Xn22WdX+9y5/PLL87Of/SwTJ07MrFmzcsQRR2TEiBH59a9/nYceeigTJkzIddddV9W/srIy559/fp599tk8/PDDqVOnTgYOHJhly5attLbZs2enV69e2WWXXTJ58uSMHTs2r732Wo444oi1+j2BmlReXp7y8vLcf//9Wbhw4Qrrn3322STJqFGjqj6zSfLuu+/moIMOyp/+9KdMnTo1ffv2Tf/+/TNz5swkybHHHpunn366WtD94osv5q9//WuOPfbY1a7vxBNPzCuvvJI///nPueeee3L99devViBx6aWX5oILLsi0adOy/fbb5+ijj86SJUuSJH/961/Tt2/fHHLIIXn++edz11135YknnsiZZ5652nVRu3z729/O2WefnenTp6dv376f2r93797ZdNNNc++991YtW7p0ae6+++6q8/eDDz5I165d8/vf/z4vvPBCTj311AwaNChPP/101TYXXnhhHnnkkfz2t7/N+PHj8+ijj2bKlCnV3mvRokW58sor85e//CX3339//vnPf7okdlVKbBR69uxZGjFiRKlUKpUWL15catmyZWnChAlV65cuXVpq27Zt6f777y+VSqXS66+/Xqpfv37pH//4R43Uy+p5+umnS0lK9913X02XwudwwgknlOrWrVtq0qRJtVejRo1KSUrz5s2r6RKhRvTq1au09957V7WXLFlSatKkSWnQoEFVy2bPnl1KUnrqqadWuo8hQ4aUDj300Kr2CSecUBowYECpVCqV3n333VKjRo1KEydOrLbN4MGDS0cffXSBR8KG4Nprry1ts802pVJp9c6dRx55pJSk9Kc//alq/bBhw0pJSi+//HLVsm9+85ulvn37rvJ9586dW0pS+utf/1oqlUqlf/7zn6UkpalTp5ZKpVJp6NChpT59+lTbZtasWaUkpRkzZqzx8cL67p577ik1b9681KhRo1LPnj1Ll1xySekvf/lL1fokpd/+9refup8dd9yxdN1111W1u3TpUvre975X1b7kkktKe+yxR1W7V69epXPOOafaPgYMGFA64YQTSqVSqTRjxoxSktKkSZOq1k+fPr2UpHTttdeutL7ln+ubbrqpav2LL75YSlKaPn16qVQqlQYNGlQ69dRTq73v448/XqpTp07p/fff/9TjZP01atSoUkVFRVV7+fmw/N/vy333u98t7bzzztWWfXRsKpVKpbPPPrv05S9/uao9bty4UoMGDUpvvfXWKt//oIMOKn3rW98qlUql0jvvvFNq0KBB6c4776xa/+abb5YaN268wnn/Uc8880wpSemdd975hCPdOJnpthGYMWNGnnnmmRx11FFJknr16uXII4/MLbfcUtVn/PjxqaysrLqZZ8uWLdOnT59qfVj/lP7/2YplZWU1XAmf1/77759p06ZVe9100001XRbUuC5dulR9Xbdu3bRo0SKdO3euWta6deskqZpBcOONN2b33XfP5ptvnvLy8owcObJqBsPHvfTSS/nggw9ywAEHVM2aKC8vz2233fapl/Oxcfss585Hz+HWrVtnk002ybbbbltt2UdnwLz88ss55phjsu2226ZZs2Zp3759kqzyPJ4yZUoeeeSRanV07Nixal+woTr00EPz6quv5ne/+1369u2bRx99NLvttttKb6OzXGVlZS666KLsuOOO2XTTTVNeXp6//e1v1T5fxx57bNXloKVSKXfcccdnmuU2ffr01KtXL7vvvnvVso4dO67WFSgf/X2xxRZbJPnP+DZlypSMHj262me9b9++WbZsWf75z3+udn3UHh89h1bXsccem0cffTSvvvpqkmTMmDE56KCD0rx58yQfznz7/ve/ny5duqRFixYpLy/P+PHjqz4DL7/8chYtWpQePXpU7XOzzTbLDjvsUO19pk6dmgEDBmSbbbZJ06ZNs99++yVZ9Vi1MfMghY3AzTffnCVLluQLX/hC1bJSqZT69etn3rx5ad68eW655Za89dZb1R6csGzZskydOjVXXnll6tatWxOl8ym22267lJWVZfr06St9+iW1R5MmTdKhQ4dqy/7973/XUDWw/vj4TYPLysqqLVv+R4dly5bl7rvvznnnnZfhw4enR48eadq0aa6++upql0x81PLL9R566KFqY2SSNGzYsMjDYAPzWc6dj5+vKzunP3rpaP/+/dO2bduMHDkyW265ZZYtW5addtopixYtWmUt/fv3z49+9KMV1i3/RztsqBo1apQDDjggBxxwQC677LKcfPLJ+e53v7vKy9wuvPDCjBs3Lj/+8Y/ToUOHNG7cOIcddli1z9cxxxyTiy++OM8991zef//9zJo1q2ryQpLUqVNnhdv0LF68uOrrz/NH8VWNb8v/+81vfjNnn332Ctut7AF51H4fv33Qp517SdKtW7d88YtfzJ133pnTTz89v/3tbzNq1Kiq9cOHD8+1116bESNGpHPnzmnSpEnOPffcqs/Ax/e/MpWVlenTp0/69OmT22+/PZtvvnlmzpyZvn37rnKs2pgJ3TZwS5YsyW233Zbhw4enT58+1dYdeuihGTNmTI4++ug88MADufPOO/OlL32pav2yZcuyzz775I9//GP69eu3rktnNWy22Wbp27dvfv7zn+fss89e4Rfz22+/7b5uwEbj8ccfT8+ePTNkyJCqZZ8002fHHXdMw4YNM3PmzKobDMPqWFvnzptvvpnp06fnF7/4RfbZZ58kyRNPPPGJ2+y222659957065du9Sr53/t2bjtuOOOVQ8nqF+/fpYuXVpt/eOPP54TTzwxAwcOTPLhPd5eeeWVan222mqr7LvvvhkzZkzef//99O7du2pWdZJsvvnm1Z48uXTp0rzwwgvZf//9kySdOnXKkiVLMnny5HTr1i3Jh1ceffzm9p/VbrvtlhdffHGFP9Ky8dh8880zZ86clEqlqlB22rRpK/Q75phjMmbMmGy11VapU6dODj744Kp1jz/+eAYMGJDjjjsuyYf/5v/HP/6RTp06JUk6dOiQ+vXrZ9KkSVVh7rx58/L3v/+9arz729/+ljfeeCM//OEP07Zt2yTJ5MmT19px13ZG5g3c73//+8ybNy+DBw9ORUVFtXWHHXZY1Sy4Fi1a5PDDD0+dOtWvOO7Xr19uvvlmodt67Prrr0/Pnj3TrVu3fO9730uXLl2yZMmSTJgwITfccEOmT59e0yUCrBMdOnTIbbfdlnHjxqV9+/b51a9+lWeffbbq8ryPa9q0aS644IKcd955WbZsWfbee+8sWLAgEydOTHl5eU444YR1fATUFmvr3GnevHlatGiRX/7yl9liiy0yc+bMXHzxxZ+4zRlnnJGRI0fm6KOPzoUXXpiWLVvmf//3f3PnnXdm5MiRrlZgg/Tmm2/m8MMPz0knnZQuXbqkadOmmTx5cq666qoMGDAgSdKuXbs8/PDD2WuvvdKwYcM0b948HTp0yH333Zf+/funrKwsQ4cOXelDSo499thcfvnlWbRoUa699tpq67785S/n/PPPz0MPPZQvfvGLufbaa6sFajvssEMOPPDAnHLKKfnlL3+ZevXq5dxzz03jxo0/1zF/+9vfTvfu3XPGGWfklFNOSZMmTTJ9+vQVHsbChmu//fbL66+/nquuuiqHHXZYxo4dmz/+8Y9p1qxZtX7HHntsrrjiinz/+9/PYYcdlkaNGlWt69ChQ+69995MnDgxzZs3zzXXXJM5c+ZUhW7l5eUZPHhwLrzwwrRo0SKtW7fOpZdeWi0n2HrrrdOgQYNcd911Oe200/LCCy/kyiuvXDffhFrIPd02cDfffHN69+69QuCWfDjTbdq0abnssssycODAFQK35X1+//vf57XXXlsX5bIG2rdvn+eeey77779/vvWtb2WnnXbKAQcckIcffjg33HBDTZcHsM6cdtppOeSQQ3LkkUdmzz33zJtvvllt1tvKXHnllbnssssybNiwdOrUKX379s2DDz64yqAOllsb506dOnVy5513ZsqUKdlpp51y3nnn5eqrr/7Ebbbccss8+eSTWbp0afr27Zuddtop55xzTioqKlb6/3awISgvL8+ee+6Za6+9Nvvuu2922mmnDB06NKecckp+9rOfJfnwMroJEyakbdu22XXXXZMk1157bZo3b56ePXumf//+6du3b3bbbbcV9n/44YfnzTffzHvvvbfCLVxOOumknHDCCTn++OPTq1evtG/fvmqW23KjRo1K27Zt06tXrxxyyCE59dRT06pVq891zF26dMljjz2Wf/zjH9lnn32y6667ZujQoS4j34h06tQp119/fX7+859n5513zjPPPJMLLrhghX7bbbdd9thjjzz//PMr3I9w6NCh2W233dK3b9/st99+adOmzQrn+NVXX5199903X/va19K7d+/svffe6dq1a9X6zTffPKNHj85vfvOb7LjjjvnhD3+YH//4x2vlmDcEZaXVuWgXAAAAAFht/vwFAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAACsNRMnTkzdunVz4IEHVlv+yiuvpKysrOpVUVGR7t2758EHH0yS7LffftXWf/zVrl27GjgaWH1CN6gFFi1aVNMlAAAArJFbbrklZ511Vp544onMnDlzhfV/+tOfMnv27Dz99NPp1q1bDj300Lzwwgu57777Mnv27MyePTvPPPNMtb6zZ8/Os88+u64PBT4ToRush/bbb7+ceeaZOf/889OyZcsccMABueaaa9K5c+c0adIkbdu2zZAhQ/Luu+9W2+7JJ59Mr169sskmm6R58+bp27dv5s2blyQplUq56qqrsu2226Zx48bZeeedc88999TE4QGwnrnnnnvSuXPnNG7cOC1atEjv3r1TWVmZJBk1alQ6deqURo0apWPHjrn++uurtjvppJPSpUuXLFy4MEmyePHidO3aNccee2yNHAcA65/KysrcfffdOf3009OvX7+MHj16hT4tWrRImzZt0rFjx3z/+9/P4sWL88gjj2SzzTZLmzZt0qZNm2y++ebV+n50GayvhG6wnrr11ltTr169PPnkk/nFL36ROnXq5Kc//WleeOGF3Hrrrfnzn/+ciy66qKr/tGnT8pWvfCVf+tKX8tRTT+WJJ55I//79s3Tp0iTJd77znYwaNSo33HBDXnzxxZx33nk57rjj8thjj9XUIQKwHpg9e3aOPvronHTSSZk+fXoeffTRHHLIISmVShk5cmQuvfTSfP/738/06dPzgx/8IEOHDs2tt96aJPnpT3+aysrKXHzxxUmSoUOH5o033qgWzAGwcbvrrruyww47ZIcddshxxx2XUaNGpVQqrbTv4sWLM3LkyCRJ/fr112WZsFaUlVZ1tgM1Zr/99sv8+fMzderUVfb5zW9+k9NPPz1vvPFGkuSYY47JzJkz88QTT6zQt7KyMi1btsyf//zn9OjRo2r5ySefnPfeey+//vWviz8IAGqF5557Ll27ds0rr7ySbbbZptq6rbfeOj/60Y9y9NFHVy377//+7/zhD3/IxIkTkyRPPfVUevXqlYsvvjjDhg3Lww8/nH333XedHgMA66+99torRxxxRM4555wsWbIkW2yxRe6444707t07r7zyStq3b5/GjRunTp06ef/997Ns2bK0a9cuU6ZMyWabbVa1n+V9p06dml122aXmDgg+g3o1XQCwcrvvvnu19iOPPJIf/OAHeemll7JgwYIsWbIkH3zwQSorK9OkSZNMmzYthx9++Er39dJLL+WDDz7IAQccUG35okWLsuuuu661YwBg/bfzzjvnK1/5Sjp37py+ffumT58+Oeyww7JkyZLMmjUrgwcPzimnnFLVf8mSJamoqKhq9+jRIxdccEGuvPLKfPvb3xa4AVBlxowZeeaZZ3LfffclSerVq5cjjzwyt9xyS3r37l3V76677krHjh3z97//Peeee25uvPHGaoEb1FZCN1hPNWnSpOrrf/3rXznooINy2mmn5corr8xmm22WJ554IoMHD87ixYuTJI0bN17lvpYtW5Ykeeihh/KFL3yh2rqGDRuuheoBqC3q1q2bCRMmZOLEiRk/fnyuu+66XHrppVVPjhs5cmT23HPPFbZZbtmyZXnyySdTt27d/OMf/1intQOwfrv55puzZMmSav8GKZVKqV+/ftW9p5Okbdu22W677bLddtulvLw8hx56aF566aW0atWqJsqGwrinG9QCkydPzpIlSzJ8+PB0794922+/fV599dVqfbp06ZKHH354pdvvuOOOadiwYWbOnJkOHTpUe7Vt23ZdHAIA67GysrLstddeueKKKzJ16tQ0aNAgTz75ZL7whS/k//2//7fC2NG+ffuqba+++upMnz49jz32WMaNG5dRo0bV4JEAsL5YsmRJbrvttgwfPjzTpk2rev3lL3/JNttskzFjxqx0u169emWnnXbK97///XVcMRTPTDeoBb74xS9myZIlue6669K/f/88+eSTufHGG6v1ueSSS9K5c+cMGTIkp512Who0aJBHHnkkhx9+eFq2bJkLLrgg5513XpYtW5a99947CxYsyMSJE1NeXp4TTjihho4MgJr29NNP5+GHH06fPn3SqlWrPP3003n99dfTqVOnXH755Tn77LPTrFmzfPWrX83ChQszefLkzJs3L+eff36mTZuWyy67LPfcc0/22muv/OQnP8k555yTXr16Zdttt63pQwOgBv3+97/PvHnzMnjw4Gq3JUiSww47LDfffHP69eu30m2/9a1v5fDDD89FF120wpU6UJuY6Qa1wC677JJrrrkmP/rRj7LTTjtlzJgxGTZsWLU+22+/fcaPH5+//OUv6datW3r06JEHHngg9ep9mK1feeWVueyyyzJs2LB06tQpffv2zYMPPlhttgIAG59mzZrlf/7nf3LQQQdl++23z3e+850MHz48X/3qV3PyySfnpptuyujRo9O5c+f06tUro0ePTvv27fPBBx/k2GOPzYknnpj+/fsnSQYPHpzevXtn0KBBVU/PBmDjdPPNN6d3794rBG5Jcuihh2batGl56623Vrptv3790q5dO7PdqPU8vRQAAAAACmamGwAAAAAUTOgGAAAAAAUTugEAAABAwYRuAAAAAFAwoRsAAAAAFEzoBgAAAAAFE7oBAAAAQMGEbgAAAABQMKEbAAAAABRM6AYAAAAABRO6AQAAAEDBhG4AAAAAULD/D7XQf97PF2XeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (race_ax, sex_ax, art_ax) = plt.subplots(1,3, sharey=True, figsize = (15, 5))\n",
    "\n",
    "race_ax.set_ylabel('residual')\n",
    "\n",
    "sns.barplot(x=data['race'],  y=res.residuals_, ax=race_ax)\n",
    "sns.barplot(x=data['sex'],  y=res.residuals_, ax=sex_ax)\n",
    "sns.barplot(x=data['ART'],  y=res.residuals_, ax=art_ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e0a1f0c-7df8-40f8-ab6f-bb2e70eb7493",
   "metadata": {},
   "source": [
    "Here we see some interesting patterns:\n",
    " - The graph of race against residuals shows us that our model is signifacntly racially biased. AA individuals are significantly 'under-estimated' by the model, C individauals are significantly over-estimated, and H individuals are significantly over-estimated.\n",
    " - The graph of sex shows that there is no real difference in the residuals. It has accounted for sex already.\n",
    " - It looks like there is a real difference across ART."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bc5658b-b99f-44f1-8746-495870be08a4",
   "metadata": {},
   "source": [
    "## _ANCOVA_"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bb494a9-d773-4f50-8c7a-52535f1684f8",
   "metadata": {},
   "source": [
    "What we have done above is create a model that _accounts_ for the effects of age, education, and YS on EDZ.\n",
    "We **subtracted** that effect (the predicted value) from the observed value thus creating the _residual_.\n",
    "This is what is \"left over\" in the observed value after accounting for covariates or nuisance variables.\n",
    "Then we plotted the _residual_ against each of our categorical variables.\n",
    "If we then took the ANOVA of these residuals we'd be testing the hypothesis:\n",
    " _When accounting for age, education, and YS is there a difference across race._\n",
    " \n",
    "This process is called an _Analysis of covariance_ or an **ANCOVA**."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b088af3-35d1-4228-a38d-0ce0edd7de10",
   "metadata": {},
   "source": [
    "### Standard first"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4c97c10-cedb-4a4a-9568-c56dfe6b737d",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": []
   },
   "source": [
    "### Q4: Perform an ANOVA between ART on the Executive Domain Z-score."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed969ccd-12ec-41b6-b6ba-cd6d7203208a",
   "metadata": {
    "deletable": false,
    "editable": false,
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "|               |    |\n",
    "| --------------|----|\n",
    "| Points        | 5  |\n",
    "| Public Checks | 4  |\n",
    "\n",
    "_Points:_ 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "0cca7821-9925-43d1-a802-62a17217125e",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a plot showing the effect of ART on EDZ\n",
    "q4_plot = sns.barplot(data = data, x = 'ART', y = 'exec_domain_z') # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "07fde2af-cad6-4b78-b88d-54d027545af9",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Source</th>\n",
       "      <th>ddof1</th>\n",
       "      <th>ddof2</th>\n",
       "      <th>F</th>\n",
       "      <th>p-unc</th>\n",
       "      <th>np2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ART</td>\n",
       "      <td>1</td>\n",
       "      <td>323</td>\n",
       "      <td>7.809699</td>\n",
       "      <td>0.005507</td>\n",
       "      <td>0.023608</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Source  ddof1  ddof2         F     p-unc       np2\n",
       "0    ART      1    323  7.809699  0.005507  0.023608"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Perform an ANOVA testing the impact of ART on EDZ\n",
    "q4_res = pg.anova(data, dv = 'exec_domain_z', between = 'ART') # SOLUTION\n",
    "q4_res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "46ef6bde-3ab5-43f9-bab2-5fc4dc400688",
   "metadata": {
    "tags": [
     "otter_assign_solution_cell"
    ]
   },
   "outputs": [],
   "source": [
    "# Does ART have a significant impact on Executive Domain? 'yes' or 'no'?\n",
    "\n",
    "q4_art_impact = 'yes' # SOLUTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e134fd24",
   "metadata": {
    "deletable": false,
    "editable": false
   },
   "outputs": [],
   "source": [
    "grader.check(\"q4_art_test\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f89b18b-531d-42a1-a96a-5f5f95449fb9",
   "metadata": {},
   "source": [
    "### With correction\n",
    "\n",
    "Nicely `pingouin` has something built right in to do this whole process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "5377a300-35e4-472b-b960-1bc8c1d59001",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Source</th>\n",
       "      <th>SS</th>\n",
       "      <th>DF</th>\n",
       "      <th>F</th>\n",
       "      <th>p-unc</th>\n",
       "      <th>np2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ART</td>\n",
       "      <td>11.879147</td>\n",
       "      <td>1</td>\n",
       "      <td>17.470083</td>\n",
       "      <td>3.770731e-05</td>\n",
       "      <td>0.051768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>79.888814</td>\n",
       "      <td>1</td>\n",
       "      <td>117.488585</td>\n",
       "      <td>1.585741e-23</td>\n",
       "      <td>0.268552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>20.033725</td>\n",
       "      <td>1</td>\n",
       "      <td>29.462623</td>\n",
       "      <td>1.128191e-07</td>\n",
       "      <td>0.084308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>17.992537</td>\n",
       "      <td>1</td>\n",
       "      <td>26.460747</td>\n",
       "      <td>4.697743e-07</td>\n",
       "      <td>0.076374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Residual</td>\n",
       "      <td>217.590675</td>\n",
       "      <td>320</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Source          SS   DF           F         p-unc       np2\n",
       "0                ART   11.879147    1   17.470083  3.770731e-05  0.051768\n",
       "1  YearsSeropositive   79.888814    1  117.488585  1.585741e-23  0.268552\n",
       "2          education   20.033725    1   29.462623  1.128191e-07  0.084308\n",
       "3                age   17.992537    1   26.460747  4.697743e-07  0.076374\n",
       "4           Residual  217.590675  320         NaN           NaN       NaN"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(x=data['ART'],  y=res.residuals_)\n",
    "\n",
    "# An ANCOVA testing the impact of ART on EDZ\n",
    "# after correcting for the impace of age, education and YS\n",
    "pg.ancova(data,\n",
    "          dv = 'exec_domain_z',\n",
    "          between = 'ART',\n",
    "          covar=['YearsSeropositive', 'education', 'age'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1409e6f5-23e5-4436-a9a6-0242f4c36c7e",
   "metadata": {},
   "source": [
    "We can notice that after correction for covaraites the F-value has increased and the p-value has decreased.\n",
    "This means the analysis is attributing more difference to race after correction and is more sure this is not due to noise."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff14833e-bda0-48a2-9c26-d2e530824231",
   "metadata": {},
   "source": [
    "The _advantage_ of using the `pg.ancova` function is that you can easily and quickly do your analysis.\n",
    "The _disadvantage_ is that you cannot examine the internal regression for Normality and Homoscedasticity."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa572f6b-0e82-4a31-ab30-4c267bfb5be0",
   "metadata": {},
   "source": [
    "But, what if we wanted to have a covariate that is a category like race?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f8a699c-8439-40c4-9728-a391a5785573",
   "metadata": {},
   "source": [
    "## Regression with categories"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89316dac-b3db-444d-9bc1-9136c1e9970c",
   "metadata": {},
   "source": [
    "So, how do you do regression with a category like race?\n",
    "\n",
    "Could it be as simple as adding it the `X` matrix?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "8fbd4b6c-dbf6-4eb2-846f-ee978ab688a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# X = data[['YearsSeropositive', 'education', 'age', 'race']]\n",
    "# y = data['processing_domain_z']\n",
    "# res = pg.linear_regression(X, y)\n",
    "# res"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6199f0af-45b8-43ef-946e-1ea31145f7a7",
   "metadata": {},
   "source": [
    "Would have been nice, but we need to get a little tricky and use _dummy_ variables.\n",
    "\n",
    "In their simplest terms, dummy variables are binary representations of categories.\n",
    "Like so."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "c2cd028f-1caf-4797-841d-0d508c7f9afd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AA</th>\n",
       "      <th>C</th>\n",
       "      <th>H</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     AA      C      H\n",
       "0  True  False  False\n",
       "1  True  False  False\n",
       "2  True  False  False\n",
       "3  True  False  False\n",
       "4  True  False  False"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.get_dummies(data['race']).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "36adb5a0-9709-402a-95e8-ec24c68524a2",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/tljh/user/lib/python3.9/site-packages/pingouin/regression.py:420: UserWarning: Design matrix supplied with `X` parameter is rank deficient (rank 6 with 7 columns). That means that one or more of the columns in `X` are a linear combination of one of more of the other columns.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>-0.194</td>\n",
       "      <td>0.294</td>\n",
       "      <td>-0.661</td>\n",
       "      <td>0.509</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.772</td>\n",
       "      <td>0.383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.046</td>\n",
       "      <td>0.003</td>\n",
       "      <td>-14.133</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.052</td>\n",
       "      <td>-0.039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>-0.054</td>\n",
       "      <td>0.019</td>\n",
       "      <td>-2.795</td>\n",
       "      <td>0.006</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.092</td>\n",
       "      <td>-0.016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.005</td>\n",
       "      <td>5.868</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.021</td>\n",
       "      <td>0.041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AA</td>\n",
       "      <td>0.410</td>\n",
       "      <td>0.104</td>\n",
       "      <td>3.941</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>C</td>\n",
       "      <td>-0.583</td>\n",
       "      <td>0.149</td>\n",
       "      <td>-3.914</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.876</td>\n",
       "      <td>-0.290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>H</td>\n",
       "      <td>-0.021</td>\n",
       "      <td>0.132</td>\n",
       "      <td>-0.162</td>\n",
       "      <td>0.871</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.282</td>\n",
       "      <td>0.239</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names   coef     se       T   pval     r2  adj_r2  CI[2.5%]  \\\n",
       "0          Intercept -0.194  0.294  -0.661  0.509  0.453   0.444    -0.772   \n",
       "1  YearsSeropositive -0.046  0.003 -14.133  0.000  0.453   0.444    -0.052   \n",
       "2          education -0.054  0.019  -2.795  0.006  0.453   0.444    -0.092   \n",
       "3                age  0.031  0.005   5.868  0.000  0.453   0.444     0.021   \n",
       "4                 AA  0.410  0.104   3.941  0.000  0.453   0.444     0.205   \n",
       "5                  C -0.583  0.149  -3.914  0.000  0.453   0.444    -0.876   \n",
       "6                  H -0.021  0.132  -0.162  0.871  0.453   0.444    -0.282   \n",
       "\n",
       "   CI[97.5%]  \n",
       "0      0.383  \n",
       "1     -0.039  \n",
       "2     -0.016  \n",
       "3      0.041  \n",
       "4      0.615  \n",
       "5     -0.290  \n",
       "6      0.239  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extracting the same continious variables\n",
    "X = data[['YearsSeropositive', 'education', 'age']]\n",
    "\n",
    "# Creating new dummy variables for race\n",
    "dummy_vals = pd.get_dummies(data['race']).astype(float)\n",
    "\n",
    "\n",
    "# Adding them the end\n",
    "X = pd.concat([X, dummy_vals], axis=1)\n",
    "\n",
    "y = data['exec_domain_z']\n",
    "\n",
    "res = pg.linear_regression(X, y)\n",
    "res.round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be9ac92a-18be-4d29-9408-9a2ae605e8fb",
   "metadata": {},
   "source": [
    "This _Warning_ is telling us that our model has fallen into the _dummy variable trap_.\n",
    "The dummy variable trap occurs when dummy variables created for categorical data in a regression model are perfectly collinear, meaning one variable can be predicted from the others, leading to redundancy.\n",
    "This happens because the inclusion of all dummy variables for a category along with a constant term (intercept) creates a situation where the sum of the dummy variables plus the intercept equals one, introducing perfect multicollinearity.\n",
    "To avoid this, one dummy variable should be dropped to serve as the reference category, ensuring the model's design matrix is full rank and the regression coefficients are estimable and interpretable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "635fc2b2-2c6e-4e54-afd5-0731a721840b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>C</th>\n",
       "      <th>H</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       C      H\n",
       "0  False  False\n",
       "1  False  False\n",
       "2  False  False\n",
       "3  False  False\n",
       "4  False  False"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.get_dummies(data['race'], drop_first=True).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "05f2d96c-2f2c-47c9-8c59-b0a068c944dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>0.216</td>\n",
       "      <td>0.381</td>\n",
       "      <td>0.567</td>\n",
       "      <td>0.571</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.534</td>\n",
       "      <td>0.966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.046</td>\n",
       "      <td>0.003</td>\n",
       "      <td>-14.133</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.052</td>\n",
       "      <td>-0.039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>-0.054</td>\n",
       "      <td>0.019</td>\n",
       "      <td>-2.795</td>\n",
       "      <td>0.006</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.092</td>\n",
       "      <td>-0.016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.005</td>\n",
       "      <td>5.868</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>0.021</td>\n",
       "      <td>0.041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C</td>\n",
       "      <td>-0.993</td>\n",
       "      <td>0.115</td>\n",
       "      <td>-8.642</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-1.219</td>\n",
       "      <td>-0.767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>H</td>\n",
       "      <td>-0.432</td>\n",
       "      <td>0.147</td>\n",
       "      <td>-2.942</td>\n",
       "      <td>0.004</td>\n",
       "      <td>0.453</td>\n",
       "      <td>0.444</td>\n",
       "      <td>-0.720</td>\n",
       "      <td>-0.143</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names   coef     se       T   pval     r2  adj_r2  CI[2.5%]  \\\n",
       "0          Intercept  0.216  0.381   0.567  0.571  0.453   0.444    -0.534   \n",
       "1  YearsSeropositive -0.046  0.003 -14.133  0.000  0.453   0.444    -0.052   \n",
       "2          education -0.054  0.019  -2.795  0.006  0.453   0.444    -0.092   \n",
       "3                age  0.031  0.005   5.868  0.000  0.453   0.444     0.021   \n",
       "4                  C -0.993  0.115  -8.642  0.000  0.453   0.444    -1.219   \n",
       "5                  H -0.432  0.147  -2.942  0.004  0.453   0.444    -0.720   \n",
       "\n",
       "   CI[97.5%]  \n",
       "0      0.966  \n",
       "1     -0.039  \n",
       "2     -0.016  \n",
       "3      0.041  \n",
       "4     -0.767  \n",
       "5     -0.143  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data[['YearsSeropositive', 'education', 'age']]\n",
    "dummy_vals = pd.get_dummies(data['race'], drop_first=True).astype(float)\n",
    "X = pd.concat([X, dummy_vals], axis=1)\n",
    "y = data['exec_domain_z']\n",
    "res = pg.linear_regression(X, y)\n",
    "res.round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72089b6c-1a01-46bc-85a7-afcc96eed850",
   "metadata": {},
   "source": [
    "We can notice a few things here:\n",
    " - **AA** has become the 'reference', the coefficients of C and H are relative to AA, which is set at 0.\n",
    "   - C individuals have a decreased score (relative to AA), which is significant.\n",
    "   - H individuals have an decreased score (relative to AA), which is significant."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89709ef9-443f-4583-b103-c825dceb39ff",
   "metadata": {},
   "source": [
    "We can look at the residuals."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "ee1f5b5d-7fcd-4edc-9d1f-0e4a91e6934d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='ART'>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (race_ax, sex_ax, art_ax) = plt.subplots(1,3, sharey=True, figsize = (15, 5))\n",
    "\n",
    "race_ax.set_ylabel('residual')\n",
    "\n",
    "sns.barplot(x=data['race'],  y=res.residuals_, ax=race_ax)\n",
    "sns.barplot(x=data['sex'],  y=res.residuals_, ax=sex_ax)\n",
    "sns.barplot(x=data['ART'],  y=res.residuals_, ax=art_ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "870e03a3-8c9d-4083-92bd-752aabd00bbc",
   "metadata": {},
   "source": [
    "Let's merge everything into a single analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "40753763-7426-47a7-87c0-8fc7bf64184d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>-0.367</td>\n",
       "      <td>0.419</td>\n",
       "      <td>-0.877</td>\n",
       "      <td>0.381</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>-1.191</td>\n",
       "      <td>0.456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.044</td>\n",
       "      <td>0.003</td>\n",
       "      <td>-13.747</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>-0.051</td>\n",
       "      <td>-0.038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>-0.060</td>\n",
       "      <td>0.019</td>\n",
       "      <td>-3.107</td>\n",
       "      <td>0.002</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>-0.098</td>\n",
       "      <td>-0.022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>0.039</td>\n",
       "      <td>0.006</td>\n",
       "      <td>6.746</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>0.028</td>\n",
       "      <td>0.051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C</td>\n",
       "      <td>-0.940</td>\n",
       "      <td>0.115</td>\n",
       "      <td>-8.189</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>-1.165</td>\n",
       "      <td>-0.714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>H</td>\n",
       "      <td>-0.382</td>\n",
       "      <td>0.146</td>\n",
       "      <td>-2.612</td>\n",
       "      <td>0.009</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>-0.670</td>\n",
       "      <td>-0.094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>male</td>\n",
       "      <td>-0.014</td>\n",
       "      <td>0.092</td>\n",
       "      <td>-0.158</td>\n",
       "      <td>0.875</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>-0.195</td>\n",
       "      <td>0.166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Truvada</td>\n",
       "      <td>0.315</td>\n",
       "      <td>0.098</td>\n",
       "      <td>3.203</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.458</td>\n",
       "      <td>0.122</td>\n",
       "      <td>0.508</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names   coef     se       T   pval    r2  adj_r2  CI[2.5%]  \\\n",
       "0          Intercept -0.367  0.419  -0.877  0.381  0.47   0.458    -1.191   \n",
       "1  YearsSeropositive -0.044  0.003 -13.747  0.000  0.47   0.458    -0.051   \n",
       "2          education -0.060  0.019  -3.107  0.002  0.47   0.458    -0.098   \n",
       "3                age  0.039  0.006   6.746  0.000  0.47   0.458     0.028   \n",
       "4                  C -0.940  0.115  -8.189  0.000  0.47   0.458    -1.165   \n",
       "5                  H -0.382  0.146  -2.612  0.009  0.47   0.458    -0.670   \n",
       "6               male -0.014  0.092  -0.158  0.875  0.47   0.458    -0.195   \n",
       "7            Truvada  0.315  0.098   3.203  0.001  0.47   0.458     0.122   \n",
       "\n",
       "   CI[97.5%]  \n",
       "0      0.456  \n",
       "1     -0.038  \n",
       "2     -0.022  \n",
       "3      0.051  \n",
       "4     -0.714  \n",
       "5     -0.094  \n",
       "6      0.166  \n",
       "7      0.508  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = pd.concat([data[['YearsSeropositive', 'education', 'age']],\n",
    "               pd.get_dummies(data['race'], drop_first=True).astype(float),\n",
    "               pd.get_dummies(data['sex'], drop_first=True).astype(float),\n",
    "               pd.get_dummies(data['ART'], drop_first=True).astype(float),\n",
    "              ], axis=1)\n",
    "y = data['exec_domain_z']\n",
    "res = pg.linear_regression(X, y)\n",
    "res.round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe67da49-98ed-43fb-b15d-c511b64757f2",
   "metadata": {},
   "source": [
    "Here our _reference_ is an AA, female taking Stavudine.\n",
    " - Everything is signifiant except for sex.\n",
    " - We see that Truvada has a _significant positive_ effect on EDZ relative to Stavudine.\n",
    "\n",
    "Since this is our final model, let's test our last normality assumption."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "46cdd616-d777-4517-979a-d51996f7f1c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='Theoretical quantiles', ylabel='Ordered quantiles'>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbcAAAGwCAYAAAAqkitTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABvkklEQVR4nO3dd1zV1R/H8dcFZQhI7gUqSZo7Rxk4ceDKkdvcIzM1JVeauTPNbZmmWZKaWxyZew/cae6NPxy4FZwgl+/vjyMXroBy8cLlwuf5eNxH3O+656Ded+f7PUOnaZqGEEIIkYbYWLoAQgghhLlJuAkhhEhzJNyEEEKkORJuQggh0hwJNyGEEGmOhJsQQog0R8JNCCFEmpPB0gVISVFRUdy4cQMXFxd0Op2liyOEEMIEmqbx6NEj8ubNi43N69tm6Srcbty4gbu7u6WLIYQQ4i1cvXoVNze31x6TrsLNxcUFUL+YzJkzW7g0QgghEmNb4GE27TxERPhzfp8ywvBd/jrpKtyib0VmzpxZwk0IIazA5t0H2XngBPYODvhW/ZDfp5Cox0rpKtyEEEJYj827D7Ju2z4A6lX3okLp9xN9rvSWFEIIkeq8Gmy1Kn9k0vkSbkIIIVKVtw02kHATQgiRipgj2EDCTQghRCphrmADCTchhBCpgDmDDSTchBBCWJi5gw0k3IQQQlhQcgQbSLgJIYSwkOQKNpBB3EIIkabo9bB7N4SEQJ48ULky2NpaulRxJWewgYSbEEKkGQEB0KcPXLsWs83NDaZNgyZNLFeuVyV3sIHclhRCiDQhIACaNTMONoDr19X2gADLlOtVKRFsIOEmhBBWT69XLTZNi7svepufnzrOklIq2EDCTQghrN7u3XFbbLFpGly9qo6zlJQMNpBwE0IIqxcSYt7jzC2lgw0k3IQQwurlyWPe48zJEsEGEm5CCGH1KldWvSITWsNTpwN3d3VcSrJUsIGEmxBCWD1bW9XdH6A2G5lLR1x5CMQE3tSpKTvezZLBBhJuQgiRJjRpAqv/fMhc2y58xkLyEwyoFt3y5Sk7zs3SwQYyiFsIIdKMBtu+Bv11LncZw+AapSwyQ0lqCDaQcBNCiLRh7Vrw94cPP+TdXwfyrgW+3VNLsIHclhRCiLTh2jXInFkFXIaUT7bUFGwg4SaEEGlD9+7wv/9BsWIp/tGpLdhAwk0IIazbiRPw/Ln6+Z13UvzjU2OwgYSbEEJYr7t3oWZNqFIFoqJS/ONTa7CBhJsQQlivnj3h9m1o2xZsUvbrPDUHG0i4CSGEdVq6VL2qVoVevVL0o1N7sIEMBRBCiFQlUStp37oFPXqAkxP88UeKttqsIdhAwk0IIVKNRK+k3bMn3LsHM2bAu++mWPmsJdhAwk0IIVKF6JW0X11wNHolbaMptAYOhBw5VPf/FGJNwQag07T41m5Nm8LCwnB1dSU0NJTMmTNbujhCCAGoW5EFCya84KhOp1pwQUEpO5VWtNQSbKZ8h0uHEiGEsLDEraStcb3NQDh9OuUKRuoJNlPJbUkhhLCwxKyQ3Ym55F8yAZ6fh1Wrkr1MYL3BBtJyE0IIi3vTCtnuBDOFr4l0coXp01OkTNYcbCDhJoQQFvf6lbQ15tAVV8Kwmf6TOjCZWXuwgYSbEEJYXOyVtF8NuC+YjS+bCSnfAJsO7ZK9LGkh2EDCTQghUoUmTVR3/3z5YrbZoMcv43TCnbOS5+/ZCTXtzCatBBtIhxIhhEg1mjSBRo1iz1Biy3sfBGJ77jTkzp2sn52Wgg0k3IQQIlWxtYVq1YBnz8DREXCBChWS9TPTWrCB3JYUQojU5+JFNap7/vxk/6i0GGwg4SaEEKlLZCS0b6+WssmUKVk/Kq0GG1hRuM2cOZNSpUqROXNmMmfOjJeXF+vXr7d0sYQQwrzGj4d9+9QabU2bJtvHpOVgAysKNzc3N8aNG8fhw4c5fPgw1atXp1GjRpw6dcrSRRNCCPP4918YPlyNZfv552T7mLQebGDlEydnzZqVCRMm0KVLl3j3h4eHEx4ebngfFhaGu7u7TJwshEh9nj2D8uXV3JFbtkCNGsnyMdYcbGl+4mS9Xs/ixYt58uQJXl5eCR43duxYXF1dDS93d/cULKUQQpjgxQsoXVot6CbB9tasquV24sQJvLy8eP78Oc7OzixcuJB69eoleLy03IQQVkevT5Z1bdJCsJnScrOqcW5FihTh2LFjPHz4kBUrVtChQwd27txJsWLF4j3e3t4ee3v7FC6lEEKY4OFD2LgRWrRQM5BIsJmFVYWbnZ0dnp6eAJQvX55Dhw4xbdo0Zs2aZeGSCSFEEvXurcaz2dqqJbfNLD0GG1jpM7domqYZ3XYUQgirsny5CraPP4bGjc1++fQabGBFLbdvv/2WunXr4u7uzqNHj1i8eDE7duxgw4YNli6aEEKYLiQEvvhCDdSePx8ymOnrOCICOnXi8MeVWXdf/c9/egs2sKJwu3XrFu3atSMkJARXV1dKlSrFhg0bqFWrlqWLJoQQptE06NIF7t+HX3+Fl49b3lpEBLRsCatWUXLZcgp99jlFurRPd8EGVhRuv//+u6WLIIQQ5rF+vXrVqwfdupnnmi9eQKtWsGoVADpNw+uDopRLh8EGVhRuQgiRZtStC/7+4OtrnjXaXrxQLbaVKwGIyJCRE+MnU86v59tf20pJuAkhREqJigIbGxVoHTqY55ovXkDr1sbB9uMkyn3dyzzXt1JW3VtSCCGsysCBasb/x4/Nc70XL+Czz2DFCvXWNgMnxk2kXN+vzHN9KyYtNyGESAkbNsCkSVC0qGq9va0XL6BNGzWcgJfB9uMkyvXr/fbXTgOk5SaEEMnt1i11G9LeHhYtevt12iIjVbAtWwbEtNjKSrAZSMtNCCGSU1SUCrbbt9UyNqVLv931IiPVWm8vgy3S1lYFW/8+Zihs2iEtNyGESE5Tpqi5Ixs0gJ5v2XsxMhLatYMlS9RbW1tO/DBBgi0eEm5CCJGcXF2hUCH444+36/YfGak6oyxerN7a2nLih/GUGfi1mQqatki4CSFEcuraVS1Amj170q+h16tbm4sWARBpY8uJMeMpM7CvmQqZ9ki4CSGEuWka/PZbTJd/O7ukXys62BYuBF4G2w/jKfONBNvrSLgJIYS5zZ2rptXq9ZYDqfV66NgR/voLiG6x/SjBlggSbkIIYU4nTqiOI1mzwujRSb+OXg+dOsGCBYAKtpOjx1JmUD8zFTRtk6EAQghhLo8fqxW1nz9Xg6vd3ZN2Hb0eOndWS+EAehsbTo4eywffDjBjYdM2abkJIYQ5aBp8+SWcPaum2apfP2nX0etVJ5R589RbGxtOjB4nwWYiCTchhDCHCxfUwGpvb/j++6RdIyoKPv9crRjAy2AbJS22pJDbkkIIYQ6FC8OBA+pZW8aMpp8fFaVabHPnAqDX2XByxBg+GDLQzAVNHyTchBDibYSFqcHZLi5Jn1orKkr1rowdbCPHUHroIDMWNH2R25JCCJFUUVFq1pCPPoKbN5N+jS++gN9/B6KD7XsJtrck4SaEEEk1fjysXg3580OOHKafHxUF3bvDnDlA9K3I7yk9dLCZC5r+SLgJIURSbN0KQ4aoYPvrL7C1Ne38qCjVu/K33wAVbKeGjaL0MAk2c5BwE0IIU127Bq1bQ4YMajybqfNGRkVBjx4we7Z6q9NxatgoSo0YkgyFTZ+kQ4kQQpiqQwe4cwd+/RU+/NC0czVNTcs1axaggu2kBJvZSbgJIYSpJk9WY9q6dTPtPE1TU3PNnAm8bLF9N5JSI75LhkKmbxJuQgiRWJqmuv2XLm16t//oFlvsYBsygpKjhiZDQYU8cxNCiMQ4cQLKlYNTp0w/V9Ogd2+YMQOIDrbhlBw9zMyFFNEk3IQQ4k0ePoQmTeDoUbh82bRzNQ369IHp0wGIIjrYhpu/nMJAwk0IIV5Hr4fPPoOLF1XX/wYNEn+upoGfH/z8M6CC7fS3wyTYUoCEmxBCvM7QobB+PXzyCYwalfjzNA2+/hp++gl4GWyDh1JizIjkKacwIuEmhBAJWbYMxo6FIkXUoqE2ifzK1DTo2xemTQOig+07SvwwMhkLK2KTcBNCiIQULQplysCqVeDqmrhzNA3694epUwEVbGcGDaHEDya0+sRbk6EAQgiRkBIl4MgR1f0/MTQNBgxQ4+BeOvPNtxQfOzqZCigSIi03IYSILTJSLRh68qR6b0qwDRwIkyYZNp0aOITi45K4cGlS6fVJX6EgDZFwE0KI2L75Rs3SP25c4s/RNHXexImGTacHfkvxH+MGm14PO3bAokXqv3r92xfZUIYVK6BkSWjRQr1PxyTchBAi2ty56pZisWKGmUTeSNNg8GCYMMGw6fSAwRT7cUycQwMCoGBB8PFRowt8fNT7gIC3KLOmwebNak25Zs3gwgUoXhwiIt7iotZPp2npJ97DwsJwdXUlNDSUzJkzW7o4QojUZO9elTaZM8PBg/Duu28+R9Pg22+NWnmn+w+iyLix7N4NISGQJw9UrqyWfWvWLG6DKvqu5/Llapy4ycLDwdMTrl9XiTlyJBQqlIQLpX6mfIdLhxIhhPjf/+DTT1XyLF+e+GAbMsQo2Ha3/4ZbFcZSu6BaFSdavnzw/Hn8dwqjp6v084NGjRK5LNyJE6qF1qQJ2NvDH39A7tzqlqQA5LakEEKoW3hZssAvv0C1am8+XtPgu+/UGLiXfiw0kCrzxtG8uXGwgWpU3bv3+stdvQq7d7/hcy9dgrZt1aTNXbpAWJjaXquWBNsrpOUmhBDvvafmjcyU6Y2H6iM1rnYeRsH5Pxi2/VhoAIMu/fjWxQgJSWDHjRswerTq6BIZCeXLww8/gIvLW39mWiXhJoRIv6ZMUc/ZPvggUcEWsEIjuNNw/B7F9IIcX6g/gy6NN0tx8uSJZ+ONG+qZ2rNn8P778P336nZkYocopFMSbkKI9GnRIjVFVpkyiRqoHRAAJ5uNYBgxA7InFOrPN5cmvOasxNHpwM1NdTwB4MkTtRJBvnyQNy907qxaa23bQgb52k4MeeYmhEh/Dh1SgeHqqkLuDcGm10NQp5EMI2YKrYnv9mOgmYIN1Gxdtroo8PdXt0ljr/I9fTp07CjBZgKrCbexY8fy4Ycf4uLiQs6cOWncuDHnzp2zdLGEENYmOBgaNlSdSJYuVZMiv+mUrqPoFzbC8H7iu30ZcHligsfHR6eDbNlUCy02N7eXwwCy74IPP4ROnSA0VI1bi4oy6TNEDKsJt507d9KzZ0/279/P5s2biYyMxNfXlydPnli6aEIIaxEWppauuXlTLUXj6/vmc0aPxsM/Zv21ye/6MeDypNecEFd062z2bLhyBbZvh4UL1X+DAkNo8ldTqFoV/v0XOnSA8+dh+PDEr0Ig4rDaQdx37twhZ86c7Ny5kypVqiTqHBnELUQ6d/WqCrTatQ2z9r/W99+r9dxemvKuH30vTzH5Y93d1cfFO0g7NFR1GClaVHVwKVfO5OunF+liEHdoaCgAWbNmTfCY8PBwwsPDDe/DoseECCHSJ3d32L8fnJ3ffOyYMUbBNtWjT6KCzd1dzZ2cI4fxDCWGwdl6Pfz2m+rG36aNeu536BAUKCA9IM3IKltumqbRqFEjHjx4wO7XjHocMWIEI0fGXRxQWm5CpDNz5qju/uXLJ+74sWPVtFovTfXozddB0157SvQMI0ZB9qpjx1RHkUOHVGvt3Dm59WgCk+6+aVaoR48eWoECBbSrV6++9rjnz59roaGhhtfVq1c1QAsNDU2hkgohLG7NGk3T6TTtvfc07cWLNx8/dqymqUlDNA20aR69Yr+N88qRQ9NWrHjDNZ880bQBAzTN1lad1KGDpt26ZY7apSuhoaGJ/g63utuSX331FWvWrGHXrl24vdrt6BX29vbY29unUMmEEKnO0aPQujU4OsKSJW/uSv/jj2qG/5d+LtiLPkE/v/aUKVPeMOHx7dvw8ccQFKRaa7/+CjVqmFAJkRRW0x7WNI1evXoREBDAtm3b8PDwsHSRhBCp2fXr0KABPH2qxrKVKfP648ePh0GDDG8PNO9J7yuvDzZQ46xfK0cOdUv022/h+HEJthRiNS23nj17snDhQlavXo2Liws3X6406+rqiqOjo4VLJ4RIVR4/VsF2/bpqWjVs+PrjJ05Ui42+dKFbD8rPmI7bPnWJ+HomxJlVJLaAANizR60Np9OpRUSls0iKspqW28yZMwkNDaVatWrkyZPH8FqyZImliyaESG3CwtQEw19+CX36vP7YSZNgwADD2wuff8l7s37B1hamvexD8mouGc0qErvzyP37aoqspk3V7cf//S/+C4hkZzUtN836OnUKISwlb161+Kij4+uDZfJk6N/f8PZi1+68N3uG4X2TJmr2kD59jJexcXOLZ9zaunXQtavq/+/lBX/+qbr3C4uwmnATQog3mjlTPd/y8nrzcjBTpkC/foa3F7t8gedvMw3v9Xq1vlp4uJruEVTfkDjj1kBdZ/JksLNTz+769k3kqqMiuUi4CSHShmXLoEcPNenw6dOv7xk5daoKoJcudumG55xfDYG2ejX89RfcuRNzipubuk0Z71qmhQqpBUT/+guKFzdXjcRbsMpB3Ekl028JkUbt2QM1a4K9vfr5datST5umRly/FFj3c571m83u3fDzz+qxWXyi724uXw5NPtVg3jxo1gycnFSPkxcvVMtNJJt0Mf2WEEIAcPas6g0ZFQUrVyYYbHo9XP76Z9772c+w7bf8Xem2fjasf/PHaJoKuJFf3eVT/y7o/l6jZhyZMkXtkGBLVSTchBDW6+ZNqFsXHjyA+fOhenWj3Xo97NihOi4W+Hs6E8N7G/b9lr8L3YJ/M+njqmrbWXCjLbobN6BWLaPhAyJ1kXATQlgvTYN33lGTHLdta3hmdv06bN2qHsM9fgw9+IWJfGU4bY57Z5OCzZZIRjKcwYwlkgwc/WwCZeb3lXkhUzEJNyGE9cqTBwIDwcGBgIC4XfYBvmQGv9DL8P4P9058fnUOkPixZ0U4R38mcolCtGIxkz4vZ0WjhNMn+eMRQlgXTVODrqNXBHF0JGCljmbN4gZbd2Yyg56G93PdO9Ll6u8kNtgy8AKA0xTnE9ZSniPccS8X/6wkIlWRcBNCWJcfflDTZX37LWgaer1aRebVft/dmMVMehjez3XvQOerf5C4YNPozwQO8hGZeALAVl0tHukyx52VRKRKEm5CCOsxbx589x14eMDy5eijdNSqBffuGR/2ObOZRXfDe3+39nS+OpfEBJsrD1nJp0xgINm5S36CATXObfnyN6wAIFINCTchhHXYsAG6dIGsWdGvXc+ImblwcoLt240P+5zZzOYLw/t5bu3odM2fxARbMU5xiA9pzGo2UYuqzv9Sx68o27erFWsk2KyHdCgRQqR+hw9D06ZoGTMyt8laepYrwvPncQ/rwhyjYJvv1pYO1/4kMcFWj39YQkucecIEh6E8/2Y4F4bayi1IKyXhJoRI/fLn536+kvS8OZTFc7ziPaQzvzOHzw3v5+drQ/tr83hTsGXNCl99BfVy5iLD4EycHLCAvoMbS6hZOQk3IUTq9XJakIA9OWl2IRAtgScpnfiD32IF24J8n9H++nziCzYnJ2jeXM3WVcD1IV5FH2JbqCBQHjoEUcLJKXnqIlKUhJsQInUKDYWGDdEPHUGfPj4JBltH5jKHrtiguksuzNeadtcXEDvYXFygdm3o3l1NfGxrC5w5A40aqQmWDx4EZ2eVfCJNkA4lQojU5/lzaNwYdu3ixuy1ccavReuAP7/TxRBsi/K2os31v4gdbCNGqNm5li2DGjVeBtuaNVChAly4oHqJODomd41ECpOWmxAiddHr1WrWO3ZAy5ZMyT0h3sPa8yd/0DlWsLXksxsLiR1sAwbA8OGxTtI0mDABBg2CTJlU3/6mTZOvLsJi3jrc9Ho9J06coECBAmTJksUcZRJCpFeaBr17w4oVUKMG+j/+5Pc8cW8wtWMec+lkCLbFeVvw2Y1FRAdb5swwZ456tmZk8GD48Udwd4e//1ZrsIk0yeTbkn5+fvz++++ACraqVatStmxZ3N3d2bFjh7nLJ4RIT2bMUK8yZSAggO8n2BMWZnxIW+bjT0dDsC3J24LWNxYTHWwtW6o12eIEG0CrVqonycGDEmxpnMnhtnz5ckq//Evx999/ExQUxNmzZ/Hz82PIkCFmL6AQIh1p2VIF0Pr1LNuYmZEjjXe3YQF/0sEQbEvzNKfVy2DLnBmWLoXFi1+ZHuv0aTh/Xv38wQeweTPkzp0StREWZHK43b17l9wv/2KsW7eO5s2bU7hwYbp06cKJEyfMXkAhRDrwRM3fSPbssGgRAXtz0aKF8XyRrVn4SrA1o2XIEkDHxIkJtNY2bAAvL2jQAMLDU6QqInUwOdxy5crF6dOn0ev1bNiwgZo1awLw9OlTbGXUoxDCVHv2qLkiN28GVH+Szz83PqQ1C5lPO2yJAmB5nqa0DFkK6MiWDfz8XmmtaRr8/DPUrw8vXsD334O9fYpUR6QOJnco6dSpEy1atCBPnjzodDpq1aoFwIEDB3j//ffNXkAhRBr233/wySfw9CnodOj1avrI+/djDmnFIqNgW5G7CS1eBhuo/idGwfbihVrYbeZMtd7bmjVQvnzK1UmkCiaH24gRIyhRogRXr16lefPm2L/8vyFbW1sGDRpk9gIKIdKoixfVyOqwMFi0iOUPa9I5Czx6FHNIC5awgLaGYAvI/SnNby4zDOh2doY4j/q7d4c//oCyZVWw5cuXQhUSqUmShgI0a9YMgOexZi7t0KGDeUokhEj7btyAWrXg1i349VcGHmnJhFeGszVnKX/RJlawNabZzeVGM5UMGBDP2mpffw0REfDrrzLjSDpm8jM3vV7P6NGjyZcvH87Ozly+fBmAoUOHGoYICCHEa339NVy5Aj/8wLKsX8QJtmYsYyGfkQE9ACtzN6LZzRVGwZYtW6xW24kTcO6c+rlECZg/X4ItnTM53MaMGYO/vz/jx4/Hzs7OsL1kyZLMmTPHrIUTQqRRM2fC9OnoBwyia1fjXc1YxiJaG4JtVe5GNL0ZEGduydmzX7batmyBSpVU5xHpESleMjnc5s2bx+zZs2nTpo1R78hSpUpx9uxZsxZOCJGGRETA8ePq56xZoWdPxvygMxqk3YQVRsG2OldDmrwSbDqdmieySRPA3x/q1lWhNm6c9IgUBiaH2/Xr1/H09IyzPSoqihcvXpilUEKINCZ6vkgvL7Xw6MtN06bFHPIpASzRtTQE25pcDfj01so4LbZhw6BZU03NiNypE7i6wrZt8LIvgBCQhHArXrw4u3fvjrN92bJllClTxiyFEkKkIZoGPXqo5laFCuqZGLB7d0yX/8asZKmuBRk0FWx/56pP41ur4gRbtmwwdCjqYdvIkeDpCfv2gbd3StZIWAGTe0sOHz6cdu3acf36daKioggICODcuXPMmzePtWvXJkcZhRDWbMgQ9YCsfHlYvRocHAD1I6hgW6Zrbgi2tTnr0/jW6njXbzM8Z2vfXo2R8/eHHDlSqCLCqmhJsGHDBq1KlSqak5OT5ujoqFWsWFHbuHFjUi6VokJDQzVACw0NtXRRhEgfJkzQNNC099/XtDt3DJtXrFCbG7JKe6GzVW9AW5uznmZDZPRbwytbNk3757frmnb2rAUrIyzNlO9wnabFnr0tbQsLC8PV1ZXQ0FAyZ85s6eIIkbaFh6vbkPfvw969apkZ1LO2nDmh4v01rNA1JaMWCcC6nHVpcPtvoojpqObgoMZhV893Dtt6tSEqSrXYZHmtdMmU73BZrFQIkTzs7dWCo3fvGoIN4LPPwPv+368EW504wQZq+bVamQ9Alfpw757qEfnOOylYCWGtEhVuWbJkQafTvflA4H7sSeGEEOnPtm0q2CpWVEEUK4z694enS/8mQNfEEGzrc9Shwe21cYINoMqT9VC9mWoF+vuDzIQkEilR4TZ16tRkLoYQIk04dAgaNQI7OwgKUktiv7RsGZydtJaVsYJtQ47afHIn/mBrxjKqTmwNDvbq3mS9eilWDWH9EhVuMm+kEOKNTp6EOnXg2TP46y+jYIuIgIVt1xm12Dbm8E0w2ABu5voAshaGuXPVszshTJCocAsLCzM8vAt7dc33V0hHDSHSoQsXoGZNePAA5s2Dhg0B1Xlk9Gg4MnodK7TG2GlqooeNOWpR/84/6F/5CtIRRR5CuEE+vp7xHrpGJ+KZGVmIN0v0M7eQkBBy5szJO++8E+/zN03T0Ol06PV6sxdSCJGKBQerYHs5wz9t26LXqzHWY8dCzcj1rNbFBNum7DWpf2ddnGDLSAT+dKQKu/hvRiD1m+SHBFp1QrxJosJt27ZtZM2aFYDt27cna4GEEFbGwUHNFfnVV/DFFyxfDm3aqFuRtdnAqljBtjl7TerdXR8n2Jx5RABNqMUWruSvQv3WcgdIvJ1EhVvVqlUNP3t4eODu7h6n9aZpGlevXjVv6YQQqZemqVmMc+ZUU2A5ODBwIIbla3zZyGpdI+y1CAC2ZK9B3XiCLSe3+If6lOcIq22b8Mnpv8DJIaVrI9IYk+eW9PDw4M6dO3G2379/Hw8PD7MUSgiRyj16BL6+sGuXeu/gwLJlMcFWi02s0TU0BNvWbD7xBlsBrrCHSpTnCDPpjn7RUmwl2IQZmBxu0c/WXvX48WMcHJL3L+WuXbto0KABefPmRafTsWrVqmT9PCFEPJ4+hQYN1DpqS5cCquNIjx5qd002s8bGONjq3NtIJBnjXCoDkTjzmBEMJ8fSGTRpLs/YhHkkeoaSvn37AqDT6Rg6dCiZMmUy7NPr9Rw4cIAPPvjA7AWM7cmTJ5QuXZpOnTrRtGnTZP0sIUQ8IiLU0jI7d0KrVoY1a3bvVhOR1GALf9s0wCFKLRq6LVu1eIPNlkj0ZOASnpTgJDMWZ6NZ8xSvjUjDEh1uR48eBVTL7cSJE0arcNvZ2VG6dGn69+9v/hLGUrduXerWrZvo48PDwwmPtTLvm4YxCCFeIzJS9RRZv1613ObNM3TT79cPqrOVtTafGIJte7Zq1L63KU6w+bCNmXxJHTZwBQ86989Gy5YpXhuRxiU63KJ7SXbq1Ilp06ZZxXi2sWPHMnLkSEsXQ4i0YdQoWL4cqldXtyMzZkSvh0qVwPXfbfxjU98QbDuzVaF2PC22RqxiCS3R0FGEczTt52F4TieEOZn8zG3u3LlWEWwAgwcPJjQ01PCS3pxCvIXevaFbN8OabMuXq1EADvu3s86mnlGw1bq3mRfYGZ3enj9ZQVMisKOp43q6LK3DxImWqIhID0xeFeDJkyeMGzeOrVu3cvv2baKiooz2X7582WyFe1v29vbY29tbuhhCWLdr18DNDbJnh1mzAAxd/quyg/U2dWOCLWvleIOtN9OYhh8PbLNx4acNrPmivEw8IpKVyeHWtWtXdu7cSbt27ciTJ0+iVwsQQlihceNgzBj1nK1SJcPMIxMmQBV2GgXbrqyVqXV/S5xgc+Uh3/AjN2zykevYZj4qUdQSNRHpjMnhtn79ev755x8qVqyYHOURQqQWv/yiFlQrWBAKFiQgAD7/XK09WoWdbLCpg2PUcwD2ZK1IrftxW2wAobxDLTYz4Wcn6pUomLJ1EOmWyc/csmTJYpiKK6U9fvyYY8eOcezYMQCCgoI4duwYwcHBFimPEGnWn39Cr16QJw9s2ULAQTeaNlXBVpldcYKtxv2tRBDzCCADL5iCH4W4CECdvsWp16OgJWoi0ivNRPPnz9eaNWumPXnyxNRT39r27ds1IM6rQ4cOiTo/NDRUA7TQ0NDkLagQ1mzZMk2zsdG07Nk17dQpLTJS07Jm1TTQtErs0p7YOKo3oO3O4q3Z8Tz6rQaa5sBTbTUNNA20BXymNWhg6QqJtMKU73CTb0tOmjSJS5cukStXLgoWLEjGjMZdff/999+3T9wEVKtWDU3Tku36QqR7er2ayt/ZGTZuhGLF+H6karFVZA8bbWqTKeoZAHuzeFHzgXGLzYUw1tCQauxkHXVZWfc31qyxVGVEemZyuDVu3DgZiiGESBVsbVWoXbkCZcuyZAmMGAHe7GWjbW0y6VWwBWbxosaDbYQTM+VeVu6xgTp8yGEW05Kl9ecRsDbuMzghUoJOS0dNobCwMFxdXQkNDbWasXpCpIidO8HODry8DJv69YPJk8GLQDbb1sJJ/xSAfVk+xufBdqNgA1hHXeqygVl044LfDCZOkb7+wrxM+Q43ueUmhEhj9u2D+vXViOygIHBxoUEDWLsWPmYfm2IF2/53KsQbbAB9mcwBKlB4wXAmtpEhQsKyTA43vV7PlClTWLp0KcHBwURERBjtv3//vtkKJ4RIZocPQ506EB4OixaBiwvly8ORIyrYNtvWxNkQbB/h89A42N7lEjZEcZH3OEtRjjUawYg2lqqMEDFMHgowcuRIJk+eTIsWLQgNDaVv3740adIEGxsbRowYkQxFFEIki+PH1Zpsjx+rYGvQgHLlVLBVYL9RsB145yN8Hu7gOY6G04tymt1UZgs1ceYRffuCrEIlUg1Tu2K+++672tq1azVN0zRnZ2ft4sWLmqZp2rRp07TWrVuberkUJUMBhHjpzBlNy5FD03Q6TT9vgbZxo6a9847qyv8R+7VHtk6Gvv0H3vlQc+SJUXf/D/hXu012TQOtB9O1hQstXSGRHpjyHW5yy+3mzZuULFkSAGdnZ0JDQwH45JNP+Oeff8yZu0KI5JIlC+TOzZHuv+Haow21a8PDh/AhB9lqWwNn/RMADrmWp9rDHTwjZv3Gj9nHdnzIyn06MpdM/XvSurWF6iFEAkwONzc3N0JCQgDw9PRk06ZNABw6dEgmKRYitYue6DxXLlYOOUz5mV14/FhtKs8ho2A77FqOqqE7jYKtMrvYTC2ceMJnLCRb346yZI1IlUwOt08//ZStW7cC0KdPH4YOHcp7771H+/bt6dy5s9kLKIQwk2vXoHx52LcPvR46d48Zg1aeQ2yzrY6LXiXdYdeycYIN4DY5uU9WmhCAW9+WTJqUojUQItHeepzb/v37CQwMxNPTk4YNG5qrXMlCxrmJdOvmTahaFc6fh/HjqbhqAIGBalc5DrPd1ueVYNvFU5wMp9vz3NBL0p7n9OrnIGuxiRRnyne4DOIWIq27exeqVYNTp4j69jsqbhvN/v1qV1mOsD2DD5kjHwFwxLUsVV4JtrbMZxTD8GE7V20KsngxNG9ugXqIdC9ZB3HPmzfvtfvbt29v6iWFEMnlwQOoVQtOneJ8w/6UnDCKiBdqVxn+ZVuG6oZg+zdzmTjB1o1ZzORLHpCFD9zvcymooCwyKqyCyS23LFmyGL1/8eIFT58+xc7OjkyZMqXqQdzSchPpTrt2sGABF2r3ovDGnwA1c0gZ/mV7Bh9cI8MAOJr5AyqH7eYJzoZTv2Yyk+nHTXLRu+gWlp4uYYkaCGFgyne4yR1KHjx4YPR6/Pgx586do1KlSixatCjJhRZCJIOJE1nvPYoiG6cRHWwfcNQo2I5lLv1KsGkMZRST6Ucw7gz32S3BJqyO2Z65HT58mLZt23L27FlzXC5ZSMtNpAvPnsGlS1CiBA0bwt9/x+wqzTF2ZqgaK9hKUTlsD49xMRyTg9ucoCSPcOHUtK006l0gpWsgRLySteWWEFtbW27cuGGuywkhkiI8HJo0gYoV6VntlFGwleI/o2D7L55gA7hDTqqzjVMzdkmwCatlcoeSNa+sPKhpGiEhIUyfPp2KFSuarWBCCBNFRKhujBs2sCtLQ2bvLGzY9WqwHXcpSeWw3YZg0xHFMEbhT0f+R0Hq9S9Ooy8tUgshzOKtFyvV6XTkyJGD6tWrM0lGdAphGS9eQKtW8PffrKcujR8sJZKMAJTkODszVOWdSDVV3nGXklR6tIdHqNs6OqKYyZd8wWxKcZy9fQNk1hFh9UwOt6jo6XuEEKlDZCR89hmsXMlGfPmUACJQU+GV4IRRsJ1wKWEUbDbomUNXOuHPIcpztMccmXVEpAmyWKkQViwiAn4b+5CaASe5RnUas8owk0hxTrIrQxWyRD4E4IRLcaNgsyWSP+lAGxayT+fFzbnrGd3B1VJVEcKsTA63vn37JvrYyZMnm3p5IUQiDRwIEyeCpmUnBzt5jLNhvbXinGRXxipkefEQgJMuxan0aC9hxITXz3xFGxZyzKUSHwWvw/Ydl/g+RgirZHK4HT16lH///ZfIyEiKFCkCwPnz57G1taVs2bKG43Q6WWZeiOTyTX897pP8KEd7DvMhd8hp2FeMU+zKWIWsLx4AcMq5GJUe7TEKNoAZ9MD7vbt8cNQfnJwQIi0xOdwaNGiAi4sLf/75p2G2kgcPHtCpUycqV65Mv379zF5IIUSMiOdRFJn0OZ2ZSyEuUZ91hn1FOc2ujJWNgq3i472E8g6gJj3Ozl2u48a3C0tSuvUyS1RBiGRn8iDufPnysWnTJooXL260/eTJk/j6+qbqsW4yiFtYvagolmXvTvMHv7GfCviyyfAM7X3OsCdjJbK9UFPgnXYuivfjQEOwOfKUVTTGk4ss6L6XYTPzWKoWQiRJsg7iDgsL49atW3G23759m0ePHpl6OSFEYmkagWV70fzBbxzkQ2qzMcFgO+P8vlGLLRNPWMsn+LKZJ54fMGxaNkvVQogUkaTFSjt16sTy5cu5du0a165dY/ny5XTp0oUmTZokRxmFEEDk92Px/m8mRyhLbTYanqEV4ewrwVYE78eBPEQ9NnDmERuoQ3W2c9WrBSVPLwE7uwQ/R4i0wORnbr/++iv9+/enbdu2vHih1s7IkCEDXbp0YYKM/BQi2fj+1ZEv+I8vmWkIrsKcMwq2s05FqBgr2Fx5yHrq4sV+otq0xd1/LmSQEUAi7UvyxMlPnjzh0qVLaJqGp6cnTlbQ20qeuQmro2lw4wb9p+SLM7i6MOfYY1eJHBF3ATjnVBivJ/t4QFbDMe9zhsMOlXFq3RB++w1ZjE1Ys2RdrDSak5MTpUqVSurpQog30TQYMgRt5ky2PtwGlDHseo/zcYLN+0mgUbA5OMCoeUVx+vAw5M8PNmabJ12IVE/+tguRWo0YAWPHcvlpbm6Q17DZkwtGwXbe6T28nwRyH9VJJBe3uFWhIY/PXKV5c6BgQQk2ke7I33ghUqNRo2DUKK7YF6ZSxDZukwuAQlxkj11FckbcAeCCk6dRsOUmhBPZq5HzwN/YLpXFg0X6JeEmRGrzww8wfDhXbAtRMXwbN1Hj0Qpxkb123uSKDrZMnng92cc9sgOQhxvssqlGjrtnYdAgGDDAYlUQwtKk25QQqYg+9DG3x87lGR5U1W/nBvkAeJdL7LWraBRs3k8DDcGWl+tsx4f3oi7AkCEwejTIFHgiHUtUuL26QOnrNGzYMMmFESI9CwiAVq2cyfFiBxmI5BrugAq2QDtvckXcBuBipkJ4Pw3kLjlenqmxnGYU5gIMHQojR0qwiXQvUeEW3wKlsUcQxJ4kWa/Xm6dkQqQjR7vNZPRvH/OCMobWGoAHl9lrX5Fc4SrYLmV695VgA9Dxb5cZeL23Cb75JoVLLkTqlKhnblFRUYbXpk2b+OCDD1i/fj0PHz4kNDSUdevWUbZsWTZs2JDc5RUizYmaPJUyv/XAn47oiFkMuCBB7LWvSO5wNd3d5UweeD3dZ1gBIB/XyMc1HByg+6wyEmxCxGLyMzc/Pz9+/fVXKlWqZNhWu3ZtMmXKRLdu3Thz5oxZCyhEmjZlCjb9+nIVN5qyAu3l/28W4AqB9t7kCb8JxA02d4LZjg9R2HBq9r/Y2spabELEZnJvyUuXLuHqGne1XldXV65cuWKOMgmR5un1cLb7FOirgs2H7VzCE4gn2BwL4vV0n2E4QH7+xw6qUYjLXKrUgcbtJNiEeJXJ4fbhhx/i5+dHSEiIYdvNmzfp168fH330kVkLJ0RatHQp9LD/nfdnxQ22/PyPQHtv8oarf19BjgXxfhYTbAW4wg6q8S5BHG85hjq7v7NYPYRIzUwOtz/++IPbt29ToEABPD098fT0JH/+/ISEhPD7778nRxmFSDMaNoSWLWGd3pftVIsTbPvsvQzBdsWxAN7PArlFbiAm2Dy4wonWP1Bq8bcWq4cQqZ3Jz9w8PT05fvw4mzdv5uzZs2iaRrFixahZs6ZRr0khhLHy5SHoyD0gG9dwpzrbDfvcCSbQwZu8z2OCzevZPsMAboAsPCAzYZxsO46S86XziBCvk6QZSnQ6Hb6+vnTr1o3evXtTq1atFAu2GTNm4OHhgYODA+XKlWP37t0p8rlCvI3y5aHKkcmcowglOW60z42rBDp4k++5WsX+f4754wQbwCfflcH1xllKSLAJ8UYmh1tUVBSjR48mX758ODs7ExQUBMDQoUOT/bbkkiVL8PPzY8iQIRw9epTKlStTt25dgoODk/VzhXgbffqoYJtMP56Sicc4G/a5cZV9Dl64Pb8OQLCju1GwvcslNlGLlb/cYPRosM2T0yJ1EMLamBxu33//Pf7+/owfPx67WKv5lixZkjlz5pi1cK+aPHkyXbp0oWvXrhQtWpSpU6fi7u7OzJkzk/VzhUiqr78G259UsAXjTjV2EMS7gBqnFujgHSfYQl6uAFCIi+ygGrXYQmMHGUMqhClMDrd58+Yxe/Zs2rRpg22shQ9LlSrF2bNnzVq42CIiIjhy5Ai+vr5G2319fQkMDIz3nPDwcMLCwoxeQqSUxo1BNzX+YMvLdfY5eOH+/BoAwQ5ueD3bZ5idxJML7KAa7lyDKVOgc2dLVUMIq2RyuF2/fh1PT88426Oionjx4oVZChWfu3fvotfryZUrl9H2XLlycfPmzXjPGTt2LK6uroaXu7t7spVPiNiWLIEtqx/Ti+kJBNvHhmC76uCG9/OYYHuP8+ygGm5ch6lTwc/PQrUQwnqZHG7FixePtxPHsmXLKFOmTDxnmNerHVc0TUuwM8vgwYMJDQ01vK5evZrs5RNCr4dOneAJzlRjR9xgc/Qif6xg83q+j+u4AaAjirUZPyUfN+Cnn9QDOyGEyUweCjB8+HDatWvH9evXiYqKIiAggHPnzjFv3jzWrl2bHGUEIHv27Nja2sZppd2+fTtOay6avb099vb2yVYmIeJz4esZFHpWmZOU5Cr5DdvzcEMF2zP1P1nXHPLh/TzQEGylSsH+/TY4nvoTjhyBL76wSPmFSAtMbrk1aNCAJUuWsG7dOnQ6HcOGDePMmTP8/fff1KpVKznKCICdnR3lypVj8+bNRts3b96Mt7d3sn2uECaZNIn3f+7JXDoBMStn5CbEKNiuO+TF+3mgYVmbtZPP89+GEBwdUeMGJNiEeCsmtdwiIyMZM2YMnTt3ZufOnclVpgT17duXdu3aUb58eby8vJg9ezbBwcF07949xcsiRByTJkH//vyP/LRgKaBul0cHW4FnasjKdYe8eD3fZ2jV/TPlPPXGV4PZrnD0KDg4WKgCQqQhmomcnJy0oKAgU08zm19++UUrUKCAZmdnp5UtW1bbuXNnos8NDQ3VAC00NDQZSyjSpYkTNQ20K+TXPLikgaaBpuUiRLvsWFCL3nDdPo+WnyuG/WM6XdC0vHnVm+nTLV0LIVI1U77DdZoWa9XRRGjcuDGNGzemY8eOyRK2ySksLAxXV1dCQ0PJnDmzpYsj0orZs+GLL/gf+fFhu6HzSC5uss/RC49nVwC4YZ8H7/BA/kdBALrVuMSsc9Xg2jWYNg1697ZM+YWwEqZ8h5vcoaRu3boMHjyYkydPUq5cOZycnIz2N2zY0NRLCmG19HroGeBLK6rSmT8MwZaTWwS+Jtiqul9m1nkfFWxTpkiwCWFmJrfcbGwS7oOi0+nQ6/VvXajkIi03YU5r596h2Zc5CA833p6TW+zL5MW7T9XUdCH2ufEK32cINoC/+hzkM39fGDYM+vZNwVILYb1M+Q43OdysmYSbMAtN40yrkWRbOoNq7OAMxQy7cnCb/Zk+NgTbTftceIXv4woehmN0Onj+HOwe3IIEhrEIIeIy5Ts8SasCRHv+/PnbnC6E9dE0ooYMpejSkYTiShgx/8BycNuoxXbTPhfe4YGGYHMnmL/5hBHdb2JnhwSbEMnI5HDT6/VGqwJcvnwZSJlVAYSwKE0jauAgbMaO4SxFqMpOwwDs7NwhMJM3hZ6qfw+37HJSMXyv0STJ2/HhE/5hWOnVFquCEOmFyeE2ZswYi60KIITFaBoXGvXHZuJ4TlGMauwwzN4fHWyeTy8BKti8IwK5TCFATbm1HR8KcRlGjZIB2kKkAKtZFUAIS1r75z3s/l7OCUrgw3ZukRuAbNwlMJM37z29CMAtuxxUjNhrCLY83GA7PrzHRRg+HIYOtVgdhEhPTB4KYKlVAYSwFL0e2n2dnXfYwSNcuEd2QAXbPicv3nuigu22XQ4qRezlEtH/PjRW0ZjCXFChNny4hWogRPpjcrhFrwpQoEABo+0ptSqAECkmKgoGDWLGo448fFiMh7F6PGblHoFO3kbBVjFiLxd5L9YFdKyoPI2Pam2B775T3SSFECnCalYFECJF6fXQtSv4+5Ofc0BMJ5DoYCv85AIAd+yyUylijyHYcnAbgGkLc9K6tRfgldKlFyLds5pVAYRIMZGR0LEj+PsTiBftmWfYlYX77HWqSJEn54GYYLtAYUB1LtlGdS7k86F1fVn5XQhLMbnlBlC7dm1q165t7rIIYXmRkdCuHSxezG4qUY91PMYFUMEW6OTN+0/OAXA3YzYqRezhPEUA9QxuKzUowSlo2RdcXCxWDSHSu7caxC1EmtOvHyxezA6qUpf1hmB7hwfsdX4l2F7EBFtW7rGVGpTiBPj5wcSJ8oxNCAtKVMstS5Ys6BL5D/X+/ftvVSAhLOnzE72pwn268ytPUZOCv8MDAp29KfpYBdu9jFmp9GIP53gfUC26LdSkNMfVBMiTJ0uwCWFhiQq3qVOnGn6+d+8e33//PbVr18bLSz0o37dvHxs3bmSojOER1ig8HG7coO/PHszZXog5zDfscuUhe50rUvSxGsN5L2MWo2ADKMQlPnC+CB16wtSpEmxCpAImT5zctGlTfHx86NWrl9H26dOns2XLFlatWmXO8pmVTJws4nj+HJo04dm+oxR9aDxzvysPCXT2ptjjM4AKtsov9hhNlAxqUv9JX16EQoUk2IRIRsk6cfLGjRupU6dOnO21a9dmy5Ytpl5OCMt5+hQaNoT169ka9iEh5DHsykwoe50rGoLtfsYsVHmx2xBsTjzmV76g1gd3mDQJ8PSUYBMiFTE53LJly8bKlSvjbF+1ahXZsmUzS6GESHaPHkHdurB5MwfzfUqTqOVEYA+8DDaXihR/fBpQwVb5xW5OUxwAR57yNw34gtls+uQni1VBCJEwk4cCjBw5ki5durBjxw7DM7f9+/ezYcMGmThZWIeHD6FOHThwgOCKram4908iyQioYNvjUpESj04B8CDDO1R5scsQbPY8ZxWN8WEHtGghU2oJkUqZHG4dO3akaNGi/PTTTwQEBKBpGsWKFWPv3r1UqFAhOcoohHk9eABXrxLVsROFF/xGJGoCcBfC2ONSiZKxgy1yF6coAUBGIlhOM3zZDJ9+CgsWQIYkDRUVQiQzk/5lvnjxgm7dujF06FD++uuv5CqTEMnLw4OIvYdwK5+b8Eh1Zz4m2E4C8DCDK1Ujd3KSkobT/qAzn/AP1K8PixdDxowWKb4Q4s1MeuaWMWPGeJ+3CZHqXbsG1arB+fMMHAj2Hnm5c0/99XfmEbszV6bUoxNATLCdoJTRJcJafQFNmsDy5RBrLUMhROpjcoeSTz/9NFV39xcijqAgqFwZdu5k7Rd/M2FCzC5nHrEncyVKhx0HIDRDZqpG7uQ4pQGwQY8LYTRrBj0WVYYVK8DBwRK1EEKYwOQHBp6enowePZrAwEDKlSuHk5OT0f7evXubrXBCvLXz56F6dbh+ncjRP9BgaD/DLtViSzjYdEQxm26U1x2hxPQt8HIdNyFE6mfyIG4PD48E9+l0Oi5fvvzWhUouMog7nTl5EmrWhFu3YOpUqq/uw/btapcTj9mTuRIfhP0HqGCrFrmDY0SvSajxCz3pwUzue35E1iObQf7OCGFRpnyHm9xyCwoKSnLBhEgxmgZt28Lt2zBrFhEdu7HdT+1y4jG7M1c2CjafyO1GwTaFr+nBTK7mLIv7wQ0SbEJYmST3Y7579y46nU4GbovUSaeDJUvg33+hdWsKuavNTjxmV+YqlAk7BkBYBhd8IrdzlLIvT9QYxyD8mEaQSyk8Tm+CLFksUgUhRNKZ1KHk4cOH9OzZk+zZs5MrVy5y5sxJ9uzZ6dWrFw8fPkymIgphgt271XM2gCJFoHVrGjZUnSUz8YSdrlUpG3YUiC/YwIMgejGdSw7F8Li4GeR/3oSwSoluud2/fx8vLy+uX79OmzZtKFq0KJqmcebMGfz9/dm6dSuBgYFkkf/LFZayaRM0bgx588Lp02Bnx6JF8Pff0cFWhXKh/wIq2KpHbuNfyhldIoh3aZ9nCyuOFIScOVO+DkIIs0h0uI0aNQo7OzsuXbpErly54uzz9fVl1KhRTJkyxeyFFOKN/v4bmjUDGxv45Rews2PZMvjsMzUX5E7XqpR/GWyPbJ2pEbmVI5Q3nN6W+fxDfbzrZ2XFWi9L1UIIYSaJvi25atUqJk6cGCfYAHLnzs348eNlgLewjGXL1ODqjBlh/XqoXZuAADX1oyNP2eVahfKhRwB4bOtEDf1WDvOh4fSv+In5tGdNtk6sXWupSgghzCnR4RYSEkLx4sUT3F+iRAlu3rxplkIJkWiLF0OrVpApE2zeDNWqEREBLVvGbrHFBFt1/TYO8ZHh9G7M4if6cI18fLR7sqVqIYQws0SHW/bs2bly5UqC+4OCgqTnpEh5ZctCsWKwfTv6j7wYNgzs7SFD5DO2v+PDh6GHgZgWW+xga8t8ZvIlN8nFgo5bsStayFK1EEKYWaLDrU6dOgwZMoSIiIg4+8LDwxk6dGi8i5gKYXaaBnfuqJ8LF0b/738MW1UWOzsYPRoceMaOd6pR4eFBAB7bZqKmfgsHiVm1ogkr8Kcj98mKX/EtDJpbxBI1EUIkk0R3KBk5ciTly5fnvffeo2fPnrz//vsAnD59mhkzZhAeHs78+fOTraBCACrY+vdXY9j27GH54YJ89pkNL16o3Q48Y/srwVZLv4UDfGx0mYe8w01y4+exhmUnS6R0LYQQycyk6beCgoLo0aMHmzZtIvo0nU5HrVq1mD59Op6enslWUHOQ6besnF4P3bvDnDlQogQjvDcxcnYew257nrPjnWp8/PAAAE9sM1FLv5l9eMe6iAboDMeHhTvIBP9CWIlkm37Lw8OD9evX8+DBAy5cuACoiZSzZs2a9NIKkRgREdC+vWqxffQRbbKuZ+HsmL939jxnexYfPn6QcLB5Ecj3fEczlvOArHzVX4JNiLQqSdNvZcmShY8++ujNBwphDk+fQvPmsG4dVKvGkJJrWPizi2G3Pc/ZlqU6Xg/2q8NtHPHVbzIKtjL8yzrq4cQTynAUpwY1jJa+EUKkLUmeW1KIFHP7Nhw5AvXr82zeMsZmdzTssuc5W7NUx/vBPuBlsEVtIpCKhmOKcYpN+OLCI1qxGMf6NVizJsVrIYRIQRJuIvUrWBACA1l52J227hmJfkpsRzhbstSgYqxgqx21kb1UMpzqyQW2UJPs3KMD/gSVa85hGagtRJpn8krcQqSIGzegdm14uT5gwLF3adIyI0+fqt12hLM1aw0qPQgEVLDVidrAHiobLmFLJH/TgDzc5Etm8KBBBw4fTvGaCCEsQMJNpD5BQVC5spoIefly9Hr4/POY3XaEsyVrTSrd3wvAMxsH6katZzdVjC6jJwNfMpO+NlOptvhLuRUpRDpiNbclx4wZwz///MOxY8ews7OTJXbSqtOnoVYt1XL7/nv0fQdQsybcv6922xHO5qy1qHx/DxATbLuoarhEVu4RSQYyu7ny7R8+VK/ug62tJSojhLAUqwm3iIgImjdvjpeXF7///ruliyOSw4EDUK+eSrKffiIg31e0fweePFG7MxLBpqy1qHJ/NwDPbeypG7WenVQzXMKVh2zCF493dWT9bwc4O6d4NYQQlmc14TZy5EgA/P39E31OeHg44eHhhvdhYWHmLpYwl8hINY4tNBT+/JMA5/Y0bRqzOyMRbM5Wi6r3Eg42Jx7zD/Upx79Q+0twckrhSgghUgurCbekGDt2rCEURSqXIQOsWAH/+x/6OvX5PNY6oRmJYFM2X6re2wWoYKsXtY4d+BiOsec5q2lERQLRt22P7fTpoNOldC2EEKlEmu5QMnjwYEJDQw2vq1evWrpI4lVz5sDL2W4oUQLq16d165hnbBl4wcZstal2byeggq1+1D9sp7rhEhmJYDnNqME2/ivSHNu5v6tFS4UQ6ZZFvwFGjBiBTqd77evwW/Tdtre3J3PmzEYvkUpoGowYobpBduxI9OC1Ro3U2qMQHWy++NzbAahg+yRqLduoYXSpUhynBls5/W59Sh9foFqBQoh0zaLfAr169aJVq1avPaZgwYIpUxiRcvR66N0bZswAT09YsAB0Ovr3x9BdPwMv2JCtNtVjBVuDqL/ZSs04lztCeXaP2Y3v18WRySKFEGDhcMuePTvZs2e3ZBFESos9AXKZMrB+PeTKRUQETH65EHYGXrAhex1q3N0OQLjOjoZRa9hCrVgX0ujOryziM+Ysc8W3WfmUr4sQItWymvs3wcHB3L9/n+DgYPR6PceOHQPUqgTO0t3benzxhQq2atVg1SpwdQXA11fdmczAC9Znr0uNu9uAl8GmrWEzvkaX+ZYfGMN3DKq0hwLN/krhSgghUjurCbdhw4bx559/Gt6XKVMGgO3bt1OtWjULlUqYbOBA1dnjl1/AwQFQz9l27lTTZa3LXo+ad7cCKtgaaavZRG2jS/RkOmP4jse53qXAEpnaXwgRl0mLlVo7WazUQoKD4flzKFw4zq5+/dTtSBVsdfG9uwWACF1GGmpr2Egdo+PbsIAFtONZljw4HtkLHh4pUgUhhOUl22KlQpjs1CmoU0e11o4fN9yGBFi0KCbY/slRD987McHWWFsVJ9gasAZ/OhLunBXH3Zsl2IQQCZLBQCL57NwJlSrBtWvQp49RsA0cCJ99Fh1s9al9ZzOggu1TbSXrqRfncnny22GbMxv2W9dD8eIpVg0hhPWRlptIHkuXQrt2qpfIokUQa8jH4sUwYQLYoGdtjk+ofWcTAC90GWiiBbCO+nEu5+AAMy7XQRceBJkypVg1hBDWSVpuwvz++ANatlSJtHGjUbD16wetW8cEW507GwEVbJ9qK/mHT4wuVZyTrKEBi399qGb2l2ATQiSCtNyE+VWqBKVLw/z56IuVZMdW2LYN/vwTrl9XwfZ3zk+oe3sDENNiezXYPLjMJnzJSwi8sxNoZIHKCCGskYSbMI/wcLUGm4eH6hX5778sD7ChSyWIvRiDCrYG1IsVbM205aylgdHl8nCDLdRUwTZ9uhovIIQQiSS3JcXbe/hQ9YisUkUFHDDgGxuaN48bbGtyNqTe7fUAROpsaaYtZ80rLbKs3GMTvrxLEFEjR0PPnilVEyFEGiHhJt7OtWtQuTLs2AHe3pAtG/37w8SJxofZoGdVrkbUv70OSDjYQGMVjSnBKc437IfN0CEpUg0hRNoi4SaS7uRJ8PJS/+3bFxYtYtkaeyZNMj5MRxQrczWmwa1/ABVszbVlrKZxPBfV8aPdMC7W+4rCqybImmxCiCSRZ24iafbuhfr11crZkyfD11+j10PXrsaH6YhiVa5GNLy1FoBIbGmpLWEVnxodZ0skdkSQwSUTAXdrYWdXCyGESCoJN5E0efKAiwvMmqW6/QNt2hg/Y1Mttk+Ng40lBND0lYtpzORLSnKCOz+vw84uawpVQgiRVkm4icSLilJ9+d3d4d131QraLyc/XrJEvaLpiCIgdxMa3VQLtEViS2sWxRNsMJqhfM4c7nt+yMdNMqZIVYQQaZs8cxOJ8+yZGn390Udw9ara9jLYli1TU2lF0xHFitxNaHxzNaCC7TMWspzmcS7bm2l8xxi0IkXIGviPag0KIcRbknATb3b7NlSvrqbUev99cHIy7AoIgBYtVKMOVLAtz92UT2MFWxv+Yhkt4ly2i+NfTMMP8uVDt3Ej5MiRErURQqQDEm7i9U6fhgoVYP9+6NhRTaeVVT0T0+vVfMjRVLA1o8nNVWo/NrThL5bS0nBM/vzw7bewc8VdfrP5ArJkUdcsUCAFKyWESOvkmZtI2O7d0KCB6hH5ww8waJBR1/wxY9QwN0VjaZ4WNAlZCahga8sCo2BzcYHLl1FzRJId3lmjbm3KDP9CCDOTcBMJy59ftaxmz1b3Hl/S61WwDR8evUVjWZ7mNAtZofZjQzvms5jWRpf74w+wvXIJcuZUSVe9egpVRAiR3ki4CWN6PQQFgaenulV49izY2xt2BwRA796q06SisSRPC6Nga888FvGZ0WUHDIBmFa5CRR81jGDPHsgoPSOFEMlDnrmJGA8fqtuQ3t4QHKy2vRJszZq9GmwtaRGyHFDB1oE/WUgbwzk6nVq/bfw396B2bdXTsmVLCTYhRLKScBPK+fPw8cewfr3qQPLOO0a7ozuPaFr0Fo3FeVvRImQZAFHo6Ig/f9HW6LyvvoKWnzxRs5mcOQPffKOm6hJCiGQk4SZgwwY1fu3cORg8GFatgsyZjQ7Zvdu488iivK1oeWMpEBNsC2gX59JNPomApk3hwAHo3BnGjk3eugghBPLMTSxcCO3agZ2d+rl163gPi5kMWWNh3ta0ihVsnZjLfNrHOcfNDSo5H1MrBjRsqKbqkomQhRApQMItvataFcqXhxkzoFy5eA9ZsgTWrgXQ+CtfG1pfV/NsRaGjM38wjw7xnjdtGth6faSafSVKQAb56yaESBnybZMe3bihOnZUqAD58qkB2gm0qJYsiW7MaSzI15bPri8CVLB14Xf+pGOcc2xsYHfflXjXrAFkhg8/TLaqCCFEfOSZW3qzZ49qqdWvr6bVggSDbeBAaNUKNE1jfr62tLm+0LCvK3Pwp1O85+32W473pKbGE04KIUQKknBLLzQNpkyBatXgzh0YNuy1czkuWwYTJgBozMvXjraxgq0Lc5hL5zjn5MgBO0btwvuXtmqQtnQeEUJYiNyWTA8ePVI9FZcvh7x51QTIFSsmeLheDz16AGj86daBdtf+Muzrym/8QZc457i6wvVNp8jo00jNorxqFZQsaf66CCFEIki4pQd9+qhgq1ZNjajOleu1h48ZA3fvavi7daT9tfmG7d2Yxe90jfecBT9eJ2PDumog+KJF4ONjxgoIIYRpJNzSgzFjoGBBNR3/G3osLl8Ow4drzHXrSIdr8wzbv+BXfqNbvOcMGACfFDoDd+/CxInqQZ0QQliQTtNi5pxI68LCwnB1dSU0NJTMrwxSTlMiIlTi1K0Ldeok6hS9HkaPhlEjNX5370Snq38a9nVnJrPoHuccnU410lpGT/x/5Yqaj1LGsgkhkoEp3+HScktrgoNVy2nfPjhxQs3n+IawCQiAbt3g3j2N3927GAXbl8yIN9gAFv0VRcug8fCwu5quq2BBM1ZECCGSTnpLpiXLl0Pp0irY2rdXI68TEWxNm6pgm+Pehc5X5xr29eAXfuXLeM8bMABa/vuNmq5r4ECzVkMIId6WhFta8Py5ano1bw4vXoC/v3plyvTa02JW0tb4zb0rXWIFW0+mM5Me8Z43fDiMd/tJPV8rVgx+/NFsVRFCCHOQ25JpQcaMcOGCmj5r0SJ4771EnRY9GfLs/N3oGvyHYXsvfmYGPeM9x80NhhZfAS391LCCDRvUgqZCCJGKSLhZK02DvXuhUiWwtVWjrjNnVhMgJ9Lq1TArfzc+D55j2PYVP/ELvRI8Z36Pfdi2a6MGaa9fD+7ub1UNIYRIDnJb0hrdvg2ffAKVK8PmzWpb9uwmBVtAALwf0I1uwb8ZtvVhKtP5Kt7jo/OzWplQdbtzxQooVeqtqiGEEMlFWm7WZtUq+OILFXD16qkOJCbS6yFs8Bd8ESvY/JjCT/RJ8JxFi9Qq3FAHgoLUlCRCCJFKScvNWjx4oNZd+/RTNZ3WtGmqN2TOnCZf6t8W3el4frbh/ddMZhp+8R6bLRsELI6g+cEBcP++2ijBJoRI5aTlZi38/WHBAvj4Y/jzTyhcOEmXudCtBx8GzDK878skpvJ1vMc2bQpLFmvYdu4K819Ow6VmUxZCiFRNwi01u38fHB3V66uv1EDp9u3VAzAT6PWqZ6Tt5B5U/numYXs/JjKFvgme16sX2P4wWgWbt7eawkQIIayA3JZMjTQNFi6EokVh1Ci1LUMG6NTJ5GALCFATh5zs1NMo2AYwnsn0i/ccnU51gqwSvEANaitUSHWtdHBIao2EECJFWUW4XblyhS5duuDh4YGjoyOFChVi+PDhREREWLpo5hcUpOaEbNMGHj+GPHmSfKmAANUJ5JsMveh1ZYZh+0B+ZCIDEjxP02D+57uw6dpZjWFbt071xhRCCCthFbclz549S1RUFLNmzcLT05OTJ0/y+eef8+TJEyZOnGjp4plHZKRaTHT4cHj2TPWEnDFDTUScBNGzj0wt+BW9gn4xbP+GcUzg9dNl+flBVa8I1XFkxYokP98TQghLsdpVASZMmMDMmTO5fPlyos9J1asCnDgBH3yglrP+6Sc1ldZbzK4/ahS4+vemT9DPhm2DGMuPDHrjudu3q6XfePwYnJ2TXAYhhDCndLEqQGhoKFmzZn3tMeHh4YSHhxveh4WFJXexTHPhgpoXsmRJ9VqyBGrUeKvprPR6tXxbZv8+RsH2LWPeGGz2PGem8wAqvz8MyCHBJoSwWlbxzO1Vly5d4ueff6Z79/iXYok2duxYXF1dDS/31DJV1IMH8PXXatLhzz+HqCi1vVmzNwabXg87dqhB1Tt2qKXbot+PGqXuYmb298Mv6CfDOUP4nrF8+4ZCafxBZzo9no7tlDRyq1cIkX5pFjR8+HANeO3r0KFDRudcv35d8/T01Lp06fLG6z9//lwLDQ01vK5evaoBWmhoaHJV6fUiIjTtp580LWtWTQNNe+89TVu9WtOiohJ1+ooVmubmpk6NftnaGr+f/K6f0YYhjDban9BrqG60+qFKFU0LD0/mX4QQQpguNDQ00d/hFn3mdvfuXe7evfvaYwoWLIjDyy7oN27cwMfHhwoVKuDv74+NjWkNT4s+c7t4ERo0gLNn1Xi1YcOgZ89EzwcZ3fPxdX9ak979mr6XpxreD2UU3zP0jdduwgpW0Aw8PODgQekZKYRIlazmmVv27NnJnsgv0uvXr+Pj40O5cuWYO3euycFmMZqmOoa4ual7ir16wYgRal6rRIru+fi6YJv4bl+jYBvOiEQFW51cR1nysD3YucDff0uwCSHSBKvoUHLjxg2qVatG/vz5mThxInfu3DHsy507twVL9hqnTsF336k11r77Tg2APn48SQOho9ddS8iEd/vT7/IUw/sRDGcUwxM8Pnt2mDoV8uWDyvpQbFs6qim9ihc3uWxCCJEaWUW4bdq0iYsXL3Lx4kXc3NyM9lnwrmr8goLUWLUFC1RTKzIypvWWxBk+QkIS3je+0AD6X5pkeD+SYYxkRILH63QwaxY0aRK9pRpcvqzWghNCiDTCKu7tdezYEU3T4n2lGrduqVuORYqouRg/+EAt5rlmzVuNV4OEJyn5sdAABlyK6dk4iqGMeE2wubnB8uXQ5FNNjRe4eVPtkGATQqQxVhFuVuHaNfjlF9UpY+lSOHwY6tR562ADtSapm5vxpcYVGsjAWMH2PUMYzkgg/s8bORKuXHnZYhs3Tt0q9fN767IJIURqJOGWVE+ewNix8O+/6n25cmpV7FOn1OwiZuzwYmurlm8DFXDjCn3DN5dilp4Zw7cMZTTxBZu7u5pBa9iwl3Mur14N336rBsT99FOc44UQIi2wimduqUpEBMyeDd9/r25F/vcfLF6s9tWsmWwf26SJuqUYPHwQfifHG7aP0w3mO+17ooPNzU2NC3/vPXU7s3LlWAsJ/PefmpDZ2Vn1jEzCQqdCCGENJNwSS6+Hv/5SnUWuXFHPqUaNStFbe+/vG0yTkz8a3ge3/oa+c8fw8T4dISHxhFlst29Dw4bw9KlqvZUsmWLlFsIUV69epV27dty+fZsMGTIwdOhQmjdvbuliCSsj4ZZYM2ZA796qx2P//jBokElj1d7W6QGDKTZxXMyGgQPJP24s6HRqkuM3CQpSt1LHjVODyYVIpTJkyMDUqVP54IMPuH37NmXLlqVevXo4OTlZumjCiki4JVbHjqrLfP/+aoBYCjr9zRDjYBswQIWUKZ1VKlRQzwPlVqRI5fLkyUOel12Ec+bMSdasWbl//76EmzCJdChJLBcXtd5aCgabXg87O39HsfE/xGzs3x9+/DHxwfbPP3Djhvo5Vy6z9N4U4m1UqVIFnU6HTqfDzs6OokWLsnDhwniPPXz4MFFRUcky6fmMGTPw8PDAwcGBcuXKsXv37tce/+jRI/z8/ChQoACOjo54e3tz6NAhk48ZMWKEof7Rr1Q7GYUVk3BLpQICYErZ76g8NybYZjv3JeDj8YkPqMOHoWlT8PWNWXlACAvSNI1jx44xceJEQkJCOHfuHHXq1KF9+/YEBQUZHXvv3j3at2/P7NmzzV6OJUuW4Ofnx5AhQzh69CiVK1embt26BAcHJ3hO165d2bx5M/Pnz+fEiRP4+vpSs2ZNrl+/btIxAMWLFyckJMTwOnHihNnrmO4l2/TNqZApM0pb0ooVmjbqve80PTrDtP2T+FrTEaXpdGr/G926pWnu7ppmY6NpmzYle5mFSIxz585pgHby5EnDthMnTmiAtn79esO258+fa5UrV9bmzZuXLOX46KOPtO7duxtte//997VBgwbFe/zTp081W1tbbe3atUbbS5curQ0ZMiTRx2iaWg2ldOnSZqhF+mPKd7i03FIZvR4ujRzGkAtjsEHNwDIFP/oxCe1ld38/P3Vcgl68gBYt4OpV9WyuVq3kL7gQiXDkyBGyZMlCsWLFALh27RpDhgzB3t6eki978GqaRseOHalevTrt2rVL8Fo//PADzs7Or33Fd6sxIiKCI0eO4Ovra7Td19eXwMDAeD8rMjISvV5vWKEkmqOjI3v27En0MdEuXLhA3rx58fDwoFWrVly+fDnBeookSv6sTT2soeW2reswoxbbFPpoEBVn/bXt219zkT591EEtWiR6rTghUkL//v01GxsbzcnJSXN0dNQAzdHRUZs7d67hmN27d2s6nU4rXbq04XX8+PE417p375524cKF176ePn0a57zr169rgLZ3716j7WPGjNEKFy6cYNm9vLy0qlWratevX9ciIyO1+fPnazqdzuicxByzbt06bfny5drx48e1zZs3a1WrVtVy5cql3b1715RfZbpkyne4hFsqcuLb4UbBNpXe8QYbaNrChQlc5PZtTcuRQ9NKltS0x49TtPxCvImPj4/21VdfaRcuXNAOHTqkVa1aNcFbgcklOtwCAwONtn///fdakSJFEjzv4sWLWpUqVTRAs7W11T788EOtTZs2WtGiRU065lWPHz/WcuXKpU2aNOntK5fGyW1JK3Ri6EiK/TDKcCvyJ77Cj6kkNFdkQpMpkyOH6kiyahVI12mRyhw9ehRvb288PT0pX748M2bMYPz48XE6kyRGUm9LZs+eHVtbW25GTxz+0u3bt8mVK1eCn1eoUCF27tzJ48ePuXr1KgcPHuTFixd4eHiYdMyrnJycKFmyJBcuXDD5dyASJuPcUoHj342ixA8jDcE216kXfk+mEV+wRa97WrnyKzvu3oXnz9XO/PmTv9BCmOjy5cs8fPiQEiVKGLYVK1YMT09PFi1axLfffmvS9bp3706LFi1ee0y+eIbu2NnZUa5cOTZv3synn35q2L5582YaNWr0xs91cnLCycmJBw8esHHjRsaPH5+kY6KFh4dz5swZKsf5Ry3eSgq0JFON1Hhbcl2nUZpeF3Mrcjo9tGxZ1a3IWJsN7+PtLfnihaZVr65pOXNq2pUrFqmHEG+ydOlSLUOGDFp4eLjR9p49e2rly5dP0bIsXrxYy5gxo/b7779rp0+f1vz8/DQnJyftSqx/Pz///LNWvXp1w/sNGzZo69ev1y5fvqxt2rRJK126tPbRRx9pERERJh3Tr18/bceOHdrly5e1/fv3a5988onm4uJi9NkifvLMLQGpLdxeDbZf+FLjZXd/0LRs2YzDzd09gWEA/fqpA5o2lQ4kItUaNGiQVqxYsTjbV61apel0Ou3q1aspWp5ffvlFK1CggGZnZ6eVLVtW27lzp9H+4cOHawUKFDC8X7Jkifbuu+9qdnZ2Wu7cubWePXtqDx8+NDonMce0bNlSy5Mnj5YxY0Ytb968WpMmTbRTp04lWz3TElO+w3WalppW/ExeYWFhuLq6EhoaSmYLL9B5fNj3lPh+GDYvf/0z6U5PfkF7Oa5ep1OTofj7qzmPE5wUeeVKtWRAsWKwf7+aSUUIIdIgU77D5ZmbBRwfMcYo2H7lC6NgA9VWu3ZNhVnr1glc6NIl6NRJdRxZvlyCTQghXpJwS2HHR/5A8VExwTaLbvRghlGwxRYS8pqL9ewJoaFqKZ6iRZOhtEIIYZ0k3FLQf6PGUmLkUGw1Nc/jb3TlS2YmGGzwmi7/AL//DsuWwWefmbmkQghh3WScWwr5b/RYSoz4zhBsUV26MjrfLNDF/0eg04G7ezxd/gEiI9V/8+VL0cVShRDCWki4JTO9Hjb3GkeJ4THBRpcu2MyexdSfYjqPxBb9furUeDqQnDkDRYrA1q3JWm4hhLBmEm7JKCAAvv/oR6rPHGIItiWZOhFQZzbY2NCkieoH8uo4Uzc3tb1Jk1cu+OQJNG+uFk0NDU2ZSgghhBWSZ26JoNfD7t2qc0eCXfJfERAA+78bz9hz32L7ci21uXSk69M5aC1sDOHVpAk0apSI62sa9OihVtP++ut4kk8IIUQ0Gef2BgEB0KeP6pYfzc0Npk1LOF/0ehhVYQLDjg4yBJs/HejC70Rha5hCKyjozSFp8Mcf0KULfPwx7NwJdnaJPFEIIdIGU77D5bbkawQEQLNmxsEGcP262h4QEP95W/oaB9uftDcEG6hG2NWrqrWWKMePq27/WbPCkiUSbEII8QYSbgnQ61WLLb52bfS2+BYNPTZ2IjWmDzYE2zza0Zk/DMEW22vHsMWWLRt89BEsWCCTIgshRCLIM7cE7N4dt8UWW+zWV7VqatvRcZMo+d0gMkSpxJtPWzoxN95ggzeMYYstXz7YsSNut0ohhBDxkpZbAhLbqoo+7uiPkyk55BtDsAVkakMn/OMNtteOYYtt8WLYsiXmJCGEEIkiLbcEJLZVlScPHB0/mZJDBhqCjc8+g8Z/EtXSFh3GtzZfO4YttnPnVAcSR0fV80TmjRRCiESTllsCKldWPRoTajBFt74yH5hCyW8HkiH64Vvr1vDnnzRpbmvaGLbYwsOhVSt4+hR++02CTQghTCQttwTY2qru/s2aqSCLr/X1U8MplBoyICbYWrWCefMgg/q1JnoM26sGDYJjx6B7d4i1UrAQQojEkXFubxDfODd3d5jWaBoNZvaLCbaWLVVvxgxv+f8L69ZB/fpQvDgcOqRuSwohhJD13MwpvtaXy6GfKDW4f0ywtWhhnmADmDMHHBxUZxIJNiGESBIJt0SwtY3p7v/vpJ8oObgfGfUvZ+Zv3lytp2aOYANYuhT+/RdKlDDP9YQQIh2SDiUmODLpJ0p+EyvYmjY1X7DduKH+myGDGrAthBAiySTcEunI5J8pOai/cbAtWgQZM779xY8ehXffhcmT3/5aQgghJNwS48iU6ZT8ph92kS/UhiZNzBdsz59D27aq+3/p0m9/PSGEEBJub7J18y7eG/ptTLB9+qnq7GGOYAMYMgROn1ZdMmvUMM81hRAinZNwe43Nuw+yNvAov7fuwgsnJ9Vt0pzBtn27uhVZtCiMHWueawohhJBwS8jm3QdZt20fACXatybjwYOqJ6O5lpsJDYUOHVQHkvnzpdu/EEKYkQwFiEfsYKtX3YtalZOh96Kjowq3TJmgXDnzX18IIdIxq2m5NWzYkPz58+Pg4ECePHlo164dN6K7z5tRigQbqBbg6NEweHDyXF8IIdIxqwk3Hx8fli5dyrlz51ixYgWXLl2iWbNmZv2MFAm2kBAYNw4iI81/bSGEEIAVzy25Zs0aGjduTHh4OBkT6OARHh5OeHi44X1YWBju7u7xzkuWIsGmaWreyPXrVceUli3N/xlCCJFGmTK3pNW03GK7f/8+f/31F97e3gkGG8DYsWNxdXU1vNzd3eM9LsVuRc6erYKtbl01H6UQQohkYVXh9s033+Dk5ES2bNkIDg5m9erVrz1+8ODBhIaGGl5Xr16Nc0yKBdvFi9C3L2TNCr//LitrCyFEMrJouI0YMQKdTvfa1+HDhw3HDxgwgKNHj7Jp0yZsbW1p3749r7uram9vT+bMmY1esaVYsEVGQvv2avHRWbMSv8y3EEKIJLHoM7e7d+9y9+7d1x5TsGBBHBwc4my/du0a7u7uBAYG4uXllajPi32/9sB/Z1Mm2AC2bIFataBdO7WYqRBCCJNZzXpu2bNnJ3v27Ek6NzqTY3cYSaxtgYfZeeAEkALBBlCzJuzYIXNHCiFECrGKQdwHDx7k4MGDVKpUiSxZsnD58mWGDRtGoUKFEt1qi23TzkPYOzgkf7BFRoKNjXpVrZp8nyOEEMKIVXQocXR0JCAggBo1alCkSBE6d+5MiRIl2LlzJ/b29km6Zoq02EaPBh+fmLXahBBCpAiraLmVLFmSbdu2vfV1om9lVipfjAql3ycsLOytr5mgY8fg++/BzU2Nb0vOzxJCiHQg+js7MV1FrHYQd1JEd0IRQghhva5evYqbm9trj0lX4RYVFcWNGzdwcXFB98o4s+jZS65evfrGXjjWIq3VKa3VB9JendJafSDt1cma66NpGo8ePSJv3rzY2Lz+qZpV3JY0FxsbmzemfXzj4axdWqtTWqsPpL06pbX6QNqrk7XWx9XVNVHHWUWHEiGEEMIUEm5CCCHSHAm3l+zt7Rk+fHiShxakRmmtTmmtPpD26pTW6gNpr05prT4JSVcdSoQQQqQP0nITQgiR5ki4CSGESHMk3IQQQqQ5Em5CCCHSHAm3BDRs2JD8+fPj4OBAnjx5aNeuHTesdALkK1eu0KVLFzw8PHB0dKRQoUIMHz6ciIgISxftrYwZMwZvb28yZcrEO++8Y+nimGzGjBl4eHjg4OBAuXLl2L17t6WLlGS7du2iQYMG5M2bF51Ox6pVqyxdpLcyduxYPvzwQ1xcXMiZMyeNGzfm3Llzli7WW5k5cyalSpUyDN728vJi/fr1li5WspFwS4CPjw9Lly7l3LlzrFixgkuXLtGsWTNLFytJzp49S1RUFLNmzeLUqVNMmTKFX3/9lW+//dbSRXsrERERNG/enC+//NLSRTHZkiVL8PPzY8iQIRw9epTKlStTt25dgoODLV20JHny5AmlS5dm+vTpli6KWezcuZOePXuyf/9+Nm/eTGRkJL6+vjx58sTSRUsyNzc3xo0bx+HDhzl8+DDVq1enUaNGnDp1ytJFSx6aSJTVq1drOp1Oi4iIsHRRzGL8+PGah4eHpYthFnPnztVcXV0tXQyTfPTRR1r37t2Ntr3//vvaoEGDLFQi8wG0lStXWroYZnX79m0N0Hbu3GnpophVlixZtDlz5li6GMlCWm6JcP/+ff766y+8vb3JmDGjpYtjFqGhoWTNmtXSxUiXIiIiOHLkCL6+vkbbfX19CQwMtFCpxOuEhoYCpJl/M3q9nsWLF/PkyZMkLfhsDSTcXuObb77BycmJbNmyERwczOrVqy1dJLO4dOkSP//8M927d7d0UdKlu3fvotfryZUrl9H2XLlycfPmTQuVSiRE0zT69u1LpUqVKFGihKWL81ZOnDiBs7Mz9vb2dO/enZUrV1KsWDFLFytZpKtwGzFiBDqd7rWvw4cPG44fMGAAR48eZdOmTdja2tK+fftELZKXUkytD8CNGzeoU6cOzZs3p2vXrhYqecKSUidr9eqyS5qmxdkmLK9Xr14cP36cRYsWWboob61IkSIcO3aM/fv38+WXX9KhQwdOnz5t6WIli3S15E2vXr1o1arVa48pWLCg4efs2bOTPXt2ChcuTNGiRXF3d2f//v2pphlvan1u3LiBj48PXl5ezJ49O5lLlzSm1skaZc+eHVtb2zittNu3b8dpzQnL+uqrr1izZg27du1643JZ1sDOzg5PT08Aypcvz6FDh5g2bRqzZs2ycMnML12FW3RYJUV0iy08PNycRXorptTn+vXr+Pj4UK5cOebOnfvGhf4s5W3+jKyFnZ0d5cqVY/PmzXz66aeG7Zs3b6ZRo0YWLJmIpmkaX331FStXrmTHjh14eHhYukjJQtO0VPWdZk7pKtwS6+DBgxw8eJBKlSqRJUsWLl++zLBhwyhUqFCqabWZ4saNG1SrVo38+fMzceJE7ty5Y9iXO3duC5bs7QQHB3P//n2Cg4PR6/UcO3YMAE9PT5ydnS1buDfo27cv7dq1o3z58oaWdHBwsNU+B338+DEXL140vA8KCuLYsWNkzZqV/PnzW7BkSdOzZ08WLlzI6tWrcXFxMbSyXV1dcXR0tHDpkubbb7+lbt26uLu78+jRIxYvXsyOHTvYsGGDpYuWPCzZVTO1On78uObj46NlzZpVs7e31woWLKh1795du3btmqWLliRz587VgHhf1qxDhw7x1mn79u2WLlqi/PLLL1qBAgU0Ozs7rWzZslbdzXz79u3x/ll06NDB0kVLkoT+vcydO9fSRUuyzp07G/6+5ciRQ6tRo4a2adMmSxcr2ciSN0IIIdKc1PngRQghhHgLEm5CCCHSHAk3IYQQaY6EmxBCiDRHwk0IIUSaI+EmhBAizZFwE0IIkeZIuAkhhEhzJNxEmnPlyhV0Op1hOi5rUbBgQaZOnWq261WrVg0/Pz+zXc8SdDodq1atAqz3z1VYhoSbsCpvWg6nY8eOli7iG/n7+/POO+/E2X7o0CG6deuW8gVKBUaMGMEHH3wQZ3tISAh169ZN+QIJqycTJwurEhISYvh5yZIlDBs2jHPnzhm2OTo68uDBA0sUDb1ej06nS/KKCzly5DBziayfNU/sLSxLWm7CquTOndvwcnV1RafTxdkW7fLly/j4+JApUyZKly7Nvn37jK4VGBhIlSpVcHR0xN3dnd69e/PkyRPD/gcPHtC+fXuyZMlCpkyZqFu3LhcuXDDsj26BrV27lmLFimFvb8///vc/IiIiGDhwIPny5cPJyYkKFSqwY8cOAHbs2EGnTp0IDQ01tDZHjBgBxL0t+fDhQ7p160auXLlwcHCgRIkSrF27FoB79+7RunVr3NzcyJQpEyVLlkzSYprjxo0jV65cuLi40KVLFwYNGmTUgorv1mbjxo2NWsgLFiygfPnyuLi4kDt3bj777DNu375t2L9jxw50Oh1bt26lfPnyZMqUCW9vb8P/lPj7+zNy5Ej+++8/w+/E398fML4tGZ/Tp09Tr149nJ2dyZUrF+3atePu3buG/cuXL6dkyZI4OjqSLVs2atasafRnLNIuCTeRZg0ZMoT+/ftz7NgxChcuTOvWrYmMjATgxIkT1K5dmyZNmnD8+HGWLFnCnj176NWrl+H8jh07cvjwYdasWcO+ffvQNI169erx4sULwzFPnz5l7NixzJkzh1OnTpEzZ046derE3r17Wbx4McePH6d58+bUqVOHCxcu4O3tzdSpU8mcOTMhISGEhITQv3//OGWPioqibt26BAYGsmDBAk6fPs24ceOwtbUF4Pnz55QrV461a9dy8uRJunXrRrt27Thw4ECifz9Lly5l+PDhjBkzhsOHD5MnTx5mzJhh8u85IiKC0aNH899//7Fq1SqCgoLivT08ZMgQJk2axOHDh8mQIQOdO3cGoGXLlvTr14/ixYsbfictW7Z84+eGhIRQtWpVPvjgAw4fPsyGDRu4desWLVq0MOxv3bo1nTt35syZM+zYsYMmTZogc8WnE5ZdlECIpJs7d67m6uoaZ3tQUJAGaHPmzDFsO3XqlAZoZ86c0TRN09q1a6d169bN6Lzdu3drNjY22rNnz7Tz589rgLZ3717D/rt372qOjo7a0qVLDZ8PaMeOHTMcc/HiRU2n02nXr183unaNGjW0wYMHv7bcBQoU0KZMmaJpmqZt3LhRs7Gx0c6dO5fo30e9evW0fv36Gd5XrVpV69OnT4LHe3l5ad27dzfaVqFCBa106dKvvUajRo1eu5TNwYMHNUB79OiRpmkxy+Fs2bLFcMw///yjAdqzZ880TdO04cOHG31uNEBbuXKlpmkxf65Hjx7VNE3Thg4dqvn6+hodf/XqVQ3Qzp07px05ckQDtCtXriRYVpF2SctNpFmlSpUy/JwnTx4Aw+2yI0eO4O/vj7Ozs+FVu3ZtoqKiCAoK4syZM2TIkIEKFSoYrpEtWzaKFCnCmTNnDNvs7OyMPufff/9F0zQKFy5sdO2dO3dy6dKlRJf92LFjuLm5Ubhw4Xj36/V6xowZQ6lSpciWLRvOzs5s2rSJ4ODgRH/GmTNn4iy+m5TFeI8ePUqjRo0oUKAALi4uVKtWDSBOWV7355EUR44cYfv27Ua/5/fffx+AS5cuUbp0aWrUqEHJkiVp3rw5v/32m8Wex4qUJx1KRJqVMWNGw886nQ5Qt/ui//vFF1/Qu3fvOOflz5+f8+fPx3tNTdMM1wLVgSX2+6ioKGxtbTly5IjhFmI0U1YHf9Nqz5MmTWLKlClMnTqVkiVL4uTkhJ+fHxEREYn+jMSwsbGJcxsv9m3ZJ0+e4Ovri6+vLwsWLCBHjhwEBwdTu3btOGV53Z9HUkRFRdGgQQN+/PHHOPvy5MmDra0tmzdvJjAwkE2bNvHzzz8zZMgQDhw4gIeHR5I/V1gHCTeRLpUtW5ZTp07h6ekZ7/5ixYoRGRnJgQMH8Pb2BlQnjvPnz1O0aNEEr1umTBn0ej23b9+mcuXK8R5jZ2eHXq9/bflKlSrFtWvXOH/+fLytt927d9OoUSPatm0LqC/6CxcuvLZsrypatCj79++nffv2hm379+83OiZHjhxGPVT1ej0nT57Ex8cHgLNnz3L37l3GjRuHu7s7AIcPH050GaIl5nfyqrJly7JixQoKFixIhgzxf5XpdDoqVqxIxYoVGTZsGAUKFGDlypX07dvX5DIK6yK3JUW69M0337Bv3z569uzJsWPHuHDhAmvWrOGrr74C4L333qNRo0Z8/vnn7Nmzh//++4+2bduSL18+GjVqlOB1CxcuTJs2bWjfvj0BAQEEBQVx6NAhfvzxR9atWweoXpGPHz9m69at3L17l6dPn8a5TtWqValSpQpNmzZl8+bNBAUFsX79ejZs2ACAp6enoVVy5swZvvjiC27evGnS76BPnz788ccf/PHHH5w/f57hw4dz6tQpo2OqV6/OP//8wz///MPZs2fp0aMHDx8+NOzPnz8/dnZ2/Pzzz1y+fJk1a9YwevRok8oR/TsJCgri2LFj3L17l/Dw8Dee07NnT+7fv0/r1q05ePAgly9fZtOmTXTu3Bm9Xs+BAwf44YcfOHz4MMHBwQQEBHDnzh2T/gdAWC8JN5EulSpVip07d3LhwgUqV65MmTJlGDp0qOFZEMDcuXMpV64cn3zyCV5eXmiaxrp164xur8Vn7ty5tG/fnn79+lGkSBEaNmzIgQMHDC0bb29vunfvTsuWLcmRIwfjx4+P9zorVqzgww8/pHXr1hQrVoyBAwcaWjdDhw6lbNmy1K5dm2rVqpE7d24aN25s0u+gZcuWDBs2jG+++YZy5crxv//9jy+//NLomM6dO9OhQwfat29P1apV8fDwMLTaQLXs/P39WbZsGcWKFWPcuHFMnDjRpHIANG3alDp16uDj40OOHDkSNawhb9687N27F71eT+3atSlRogR9+vTB1dUVGxsbMmfOzK5du6hXrx6FCxfmu+++Y9KkSTIoPJ3Qaa/eUBdCpFsjRoxg1apVMsWVsHrSchNCCJHmSLgJIYRIc+S2pBBCiDRHWm5CCCHSHAk3IYQQaY6EmxBCiDRHwk0IIUSaI+EmhBAizZFwE0IIkeZIuAkhhEhzJNyEEEKkOf8Hke4e7syFlgcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pg.qqplot(res.residuals_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "bb8dcb61-82af-49a9-a923-e4c58a0a220b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>W</th>\n",
       "      <th>pval</th>\n",
       "      <th>normal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.832024</td>\n",
       "      <td>0.659672</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          W      pval  normal\n",
       "0  0.832024  0.659672    True"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pg.normality(res.residuals_, method='normaltest')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77d9739b-d623-40f1-ade2-3ab1b755d7b2",
   "metadata": {},
   "source": [
    "Perfect, now we know that our final model passes the _Normal Distribution of Errors_ assumption."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63741a0f-627f-4981-b5c0-ef8b302d3335",
   "metadata": {},
   "source": [
    "What about understanding which parameters have the largest impact on the model?\n",
    "Stated another way: which features are most important to determing EDZ?\n",
    "\n",
    "Nicely, `pingouin` can do this for us."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "871beb97-cdcc-44ae-bb13-4ed78f36d495",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "      <th>relimp</th>\n",
       "      <th>relimp_perc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Intercept</td>\n",
       "      <td>-0.367108</td>\n",
       "      <td>0.418546</td>\n",
       "      <td>-0.877105</td>\n",
       "      <td>3.810941e-01</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-1.190587</td>\n",
       "      <td>0.456370</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.044294</td>\n",
       "      <td>0.003222</td>\n",
       "      <td>-13.746688</td>\n",
       "      <td>4.748977e-34</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.050633</td>\n",
       "      <td>-0.037954</td>\n",
       "      <td>0.275883</td>\n",
       "      <td>58.718414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>-0.059910</td>\n",
       "      <td>0.019281</td>\n",
       "      <td>-3.107223</td>\n",
       "      <td>2.059458e-03</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.097844</td>\n",
       "      <td>-0.021975</td>\n",
       "      <td>0.039358</td>\n",
       "      <td>8.376948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>0.039215</td>\n",
       "      <td>0.005813</td>\n",
       "      <td>6.745778</td>\n",
       "      <td>7.231020e-11</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>0.027777</td>\n",
       "      <td>0.050652</td>\n",
       "      <td>0.039614</td>\n",
       "      <td>8.431478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C</td>\n",
       "      <td>-0.939704</td>\n",
       "      <td>0.114749</td>\n",
       "      <td>-8.189228</td>\n",
       "      <td>6.513749e-15</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-1.165470</td>\n",
       "      <td>-0.713939</td>\n",
       "      <td>0.075652</td>\n",
       "      <td>16.101683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>H</td>\n",
       "      <td>-0.382354</td>\n",
       "      <td>0.146409</td>\n",
       "      <td>-2.611538</td>\n",
       "      <td>9.442348e-03</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.670411</td>\n",
       "      <td>-0.094297</td>\n",
       "      <td>0.015979</td>\n",
       "      <td>3.400943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>male</td>\n",
       "      <td>-0.014446</td>\n",
       "      <td>0.091578</td>\n",
       "      <td>-0.157748</td>\n",
       "      <td>8.747561e-01</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.194624</td>\n",
       "      <td>0.165732</td>\n",
       "      <td>0.000484</td>\n",
       "      <td>0.102939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Truvada</td>\n",
       "      <td>0.314984</td>\n",
       "      <td>0.098327</td>\n",
       "      <td>3.203452</td>\n",
       "      <td>1.495929e-03</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>0.121529</td>\n",
       "      <td>0.508440</td>\n",
       "      <td>0.022870</td>\n",
       "      <td>4.867595</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names      coef        se          T          pval       r2  \\\n",
       "0          Intercept -0.367108  0.418546  -0.877105  3.810941e-01  0.46984   \n",
       "1  YearsSeropositive -0.044294  0.003222 -13.746688  4.748977e-34  0.46984   \n",
       "2          education -0.059910  0.019281  -3.107223  2.059458e-03  0.46984   \n",
       "3                age  0.039215  0.005813   6.745778  7.231020e-11  0.46984   \n",
       "4                  C -0.939704  0.114749  -8.189228  6.513749e-15  0.46984   \n",
       "5                  H -0.382354  0.146409  -2.611538  9.442348e-03  0.46984   \n",
       "6               male -0.014446  0.091578  -0.157748  8.747561e-01  0.46984   \n",
       "7            Truvada  0.314984  0.098327   3.203452  1.495929e-03  0.46984   \n",
       "\n",
       "     adj_r2  CI[2.5%]  CI[97.5%]    relimp  relimp_perc  \n",
       "0  0.458133 -1.190587   0.456370       NaN          NaN  \n",
       "1  0.458133 -0.050633  -0.037954  0.275883    58.718414  \n",
       "2  0.458133 -0.097844  -0.021975  0.039358     8.376948  \n",
       "3  0.458133  0.027777   0.050652  0.039614     8.431478  \n",
       "4  0.458133 -1.165470  -0.713939  0.075652    16.101683  \n",
       "5  0.458133 -0.670411  -0.094297  0.015979     3.400943  \n",
       "6  0.458133 -0.194624   0.165732  0.000484     0.102939  \n",
       "7  0.458133  0.121529   0.508440  0.022870     4.867595  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_with_imp = pg.linear_regression(X, y, relimp=True)\n",
    "res_with_imp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "1a5030e3-b8b5-4918-8939-381a5bc28592",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>names</th>\n",
       "      <th>coef</th>\n",
       "      <th>se</th>\n",
       "      <th>T</th>\n",
       "      <th>pval</th>\n",
       "      <th>r2</th>\n",
       "      <th>adj_r2</th>\n",
       "      <th>CI[2.5%]</th>\n",
       "      <th>CI[97.5%]</th>\n",
       "      <th>relimp</th>\n",
       "      <th>relimp_perc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>YearsSeropositive</td>\n",
       "      <td>-0.044294</td>\n",
       "      <td>0.003222</td>\n",
       "      <td>-13.746688</td>\n",
       "      <td>4.748977e-34</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.050633</td>\n",
       "      <td>-0.037954</td>\n",
       "      <td>0.275883</td>\n",
       "      <td>58.718414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>C</td>\n",
       "      <td>-0.939704</td>\n",
       "      <td>0.114749</td>\n",
       "      <td>-8.189228</td>\n",
       "      <td>6.513749e-15</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-1.165470</td>\n",
       "      <td>-0.713939</td>\n",
       "      <td>0.075652</td>\n",
       "      <td>16.101683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>age</td>\n",
       "      <td>0.039215</td>\n",
       "      <td>0.005813</td>\n",
       "      <td>6.745778</td>\n",
       "      <td>7.231020e-11</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>0.027777</td>\n",
       "      <td>0.050652</td>\n",
       "      <td>0.039614</td>\n",
       "      <td>8.431478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>education</td>\n",
       "      <td>-0.059910</td>\n",
       "      <td>0.019281</td>\n",
       "      <td>-3.107223</td>\n",
       "      <td>2.059458e-03</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.097844</td>\n",
       "      <td>-0.021975</td>\n",
       "      <td>0.039358</td>\n",
       "      <td>8.376948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Truvada</td>\n",
       "      <td>0.314984</td>\n",
       "      <td>0.098327</td>\n",
       "      <td>3.203452</td>\n",
       "      <td>1.495929e-03</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>0.121529</td>\n",
       "      <td>0.508440</td>\n",
       "      <td>0.022870</td>\n",
       "      <td>4.867595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>H</td>\n",
       "      <td>-0.382354</td>\n",
       "      <td>0.146409</td>\n",
       "      <td>-2.611538</td>\n",
       "      <td>9.442348e-03</td>\n",
       "      <td>0.46984</td>\n",
       "      <td>0.458133</td>\n",
       "      <td>-0.670411</td>\n",
       "      <td>-0.094297</td>\n",
       "      <td>0.015979</td>\n",
       "      <td>3.400943</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               names      coef        se          T          pval       r2  \\\n",
       "1  YearsSeropositive -0.044294  0.003222 -13.746688  4.748977e-34  0.46984   \n",
       "4                  C -0.939704  0.114749  -8.189228  6.513749e-15  0.46984   \n",
       "3                age  0.039215  0.005813   6.745778  7.231020e-11  0.46984   \n",
       "2          education -0.059910  0.019281  -3.107223  2.059458e-03  0.46984   \n",
       "7            Truvada  0.314984  0.098327   3.203452  1.495929e-03  0.46984   \n",
       "5                  H -0.382354  0.146409  -2.611538  9.442348e-03  0.46984   \n",
       "\n",
       "     adj_r2  CI[2.5%]  CI[97.5%]    relimp  relimp_perc  \n",
       "1  0.458133 -0.050633  -0.037954  0.275883    58.718414  \n",
       "4  0.458133 -1.165470  -0.713939  0.075652    16.101683  \n",
       "3  0.458133  0.027777   0.050652  0.039614     8.431478  \n",
       "2  0.458133 -0.097844  -0.021975  0.039358     8.376948  \n",
       "7  0.458133  0.121529   0.508440  0.022870     4.867595  \n",
       "5  0.458133 -0.670411  -0.094297  0.015979     3.400943  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# After filtering and sorting\n",
    "res_with_imp.query('pval<0.01').sort_values('relimp_perc', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dea90faa-7e62-470e-8b38-bc4ec6c4b94d",
   "metadata": {},
   "source": [
    "## Over fitting"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34122ab1-a41f-40ae-8404-13952ec40432",
   "metadata": {},
   "source": [
    "In principle we can continue to add more and more variables to the `X` and just let the computer figure out the p-value of each.\n",
    "\n",
    "There are a few reasons we shouldn't take this tack.\n",
    " - **Overfitting** : A larger model will **ALWAYS** fit better than a smaller model. This doesn't mean the larger model is **better** at predicting _all samples_, it just means it fits **these** samples better.\n",
    " - **Explainability** : Large models with many parameters are difficult to explain and reason about. We are biologists, not data scientists. Our job is to reason about the _result_ of the analysis, not create the best fitting model.\n",
    " - **Statistical power** : As you add more noise features you lose the power to detect real features.\n",
    "\n",
    "So, you should limit yourself to only those features that you think are biologically meaningful."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f85001ad-e7d5-4fa1-acb4-bf831e249167",
   "metadata": {},
   "source": [
    "When planning experiments there are a couple of things you can do to avoid overfitting:\n",
    " - **Sample size** : While there is no strict rule, you should plan to have _at least_ 10 samples per feature in your model.\n",
    " - **Even sampling** : It is ideal to have a roughly equal representation of the entire parameter space. If you have categories, you should have an equal number of each. If you have continious data, you should have both high and low values. If you have many parameters, you should have an equal number of each of their interactions as well.\n",
    "\n",
    "These are good guidelines for all model-fitting style analyses."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "c7b277ae-b218-400b-bf21-2dbe1d4dfd72",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: 7\n",
      "Obs: 325\n"
     ]
    }
   ],
   "source": [
    "print('Features:', len(X.columns))\n",
    "print('Obs:', len(X.index))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a555f8e6-5863-4b26-bff3-8cef65f03861",
   "metadata": {},
   "source": [
    "## Even more regression"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "877c659e-f08a-4108-bdd9-6a4c1144fed9",
   "metadata": {},
   "source": [
    "There are a number of regression based tools in `pingouin` that we didn't cover that may be useful to explore.\n",
    " - `pg.logistic_regression` : This works similar to linear regression but is for binary dependent variables.\n",
    "Each feature is regressed to create an equation that estimates the likelihood of the `dv` being `True`.\n",
    " - `pg.partial_corr` : Like the ANCOVA, this is a tool for removing the effect of covariates and then calculating a correlation coefficient.\n",
    " - `pg.rm_corr` : Correlation with repeated measures. This is useful if you have measured the same _sample_ multiple times and want to account for intermeasurment variability.\n",
    " - `pg.mediation_analysis` : Tests the hypothesis that the independent variable `X` influences the dependent variable `Y` by a change in mediator `M`; like so `X -> M -> Y`.\n",
    "This is useful to disentangle causal effects from covariation."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01aa3342",
   "metadata": {},
   "source": [
    "---------------------------------------------"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a34057e2-823a-4c03-a124-ff10e5db70ea",
   "metadata": {
    "tags": [
     "remove_cell"
    ]
   },
   "source": [
    "## Submission\n",
    "\n",
    "You do not need to submit this walkthrough notebook.\n",
    "Simply complete the quiz."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "74b8cf4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grader.check_all()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "otter": {
   "assignment_name": "Module09_walkthrough"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}