diff --git a/Rethinking_2/Chp_07.ipynb b/Rethinking_2/Chp_07.ipynb
new file mode 100644
index 0000000..9d75b71
--- /dev/null
+++ b/Rethinking_2/Chp_07.ipynb
@@ -0,0 +1,2650 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Chapter 7"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import arviz as az\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import pymc3 as pm\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "from patsy import dmatrix\n",
+    "from scipy import stats\n",
+    "from scipy.special import logsumexp"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%config Inline.figure_format = 'retina'\n",
+    "az.style.use('arviz-darkgrid')\n",
+    "az.rcParams['stats.credible_interval'] = 0.89  # set credible interval for entire notebook\n",
+    "np.random.seed(0)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "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>species</th>\n",
+       "      <th>brain</th>\n",
+       "      <th>mass</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>afarensis</td>\n",
+       "      <td>438</td>\n",
+       "      <td>37.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>africanus</td>\n",
+       "      <td>452</td>\n",
+       "      <td>35.5</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>habilis</td>\n",
+       "      <td>612</td>\n",
+       "      <td>34.5</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>boisei</td>\n",
+       "      <td>521</td>\n",
+       "      <td>41.5</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>rudolfensis</td>\n",
+       "      <td>752</td>\n",
+       "      <td>55.5</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>ergaster</td>\n",
+       "      <td>871</td>\n",
+       "      <td>61.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>sapiens</td>\n",
+       "      <td>1350</td>\n",
+       "      <td>53.5</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       species  brain  mass\n",
+       "0    afarensis    438  37.0\n",
+       "1    africanus    452  35.5\n",
+       "2      habilis    612  34.5\n",
+       "3       boisei    521  41.5\n",
+       "4  rudolfensis    752  55.5\n",
+       "5     ergaster    871  61.0\n",
+       "6      sapiens   1350  53.5"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "brains = pd.DataFrame.from_dict(\n",
+    "    {\n",
+    "        \"species\": [\n",
+    "            \"afarensis\",\n",
+    "            \"africanus\",\n",
+    "            \"habilis\",\n",
+    "            \"boisei\",\n",
+    "            \"rudolfensis\",\n",
+    "            \"ergaster\",\n",
+    "            \"sapiens\",\n",
+    "        ],\n",
+    "        \"brain\": [438, 452, 612, 521, 752, 871, 1350],  # volume in cc\n",
+    "        \"mass\": [37.0, 35.5, 34.5, 41.5, 55.5, 61.0, 53.5],  # mass in kg\n",
+    "    }\n",
+    ")\n",
+    "\n",
+    "brains"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Figure 7.2\n",
+    "\n",
+    "plt.scatter(brains.mass, brains.brain)\n",
+    "\n",
+    "# point labels\n",
+    "for i, r in brains.iterrows():\n",
+    "    if r.species == \"afarensis\":\n",
+    "        plt.text(r.mass + 0.5, r.brain, r.species, ha=\"left\", va=\"center\")\n",
+    "    elif r.species == \"sapiens\":\n",
+    "        plt.text(r.mass, r.brain - 25, r.species, ha=\"center\", va=\"top\")\n",
+    "    else:\n",
+    "        plt.text(r.mass, r.brain + 25, r.species, ha=\"center\")\n",
+    "\n",
+    "plt.xlabel(\"body mass (kg)\")\n",
+    "plt.ylabel(\"brain volume (cc)\");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "brains.loc[:, \"mass_std\"] = (\n",
+    "    brains.loc[:, \"mass\"] - brains.loc[:, \"mass\"].mean()\n",
+    ") / brains.loc[:, \"mass\"].std()\n",
+    "brains.loc[:, \"brain_std\"] = brains.loc[:, \"brain\"] / brains.loc[:, \"brain\"].max()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.3\n",
+    "\n",
+    "This is modified from [Chapter 6 of 1st Edition](https://nbviewer.jupyter.org/github/pymc-devs/resources/blob/master/Rethinking/Chp_06.ipynb) (6.2 - 6.6)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/statsmodels/stats/stattools.py:71: ValueWarning: omni_normtest is not valid with less than 8 observations; 7 samples were given.\n",
+      "  \"samples were given.\" % int(n), ValueWarning)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>        <td>brain_std</td>    <th>  R-squared:         </th> <td>   0.490</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.388</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   4.807</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Tue, 19 May 2020</td> <th>  Prob (F-statistic):</th>  <td>0.0798</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>16:36:51</td>     <th>  Log-Likelihood:    </th> <td>  2.9925</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>     7</td>      <th>  AIC:               </th> <td>  -1.985</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>     5</td>      <th>  BIC:               </th> <td>  -2.093</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th> <td>    0.5287</td> <td>    0.071</td> <td>    7.492</td> <td> 0.001</td> <td>    0.347</td> <td>    0.710</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>mass_std</th>  <td>    0.1671</td> <td>    0.076</td> <td>    2.192</td> <td> 0.080</td> <td>   -0.029</td> <td>    0.363</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>   nan</td> <th>  Durbin-Watson:     </th> <td>   1.561</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td>   nan</td> <th>  Jarque-Bera (JB):  </th> <td>   2.372</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 1.399</td> <th>  Prob(JB):          </th> <td>   0.305</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 3.548</td> <th>  Cond. No.          </th> <td>    1.08</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:              brain_std   R-squared:                       0.490\n",
+       "Model:                            OLS   Adj. R-squared:                  0.388\n",
+       "Method:                 Least Squares   F-statistic:                     4.807\n",
+       "Date:                Tue, 19 May 2020   Prob (F-statistic):             0.0798\n",
+       "Time:                        16:36:51   Log-Likelihood:                 2.9925\n",
+       "No. Observations:                   7   AIC:                            -1.985\n",
+       "Df Residuals:                       5   BIC:                            -2.093\n",
+       "Df Model:                           1                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "==============================================================================\n",
+       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------\n",
+       "Intercept      0.5287      0.071      7.492      0.001       0.347       0.710\n",
+       "mass_std       0.1671      0.076      2.192      0.080      -0.029       0.363\n",
+       "==============================================================================\n",
+       "Omnibus:                          nan   Durbin-Watson:                   1.561\n",
+       "Prob(Omnibus):                    nan   Jarque-Bera (JB):                2.372\n",
+       "Skew:                           1.399   Prob(JB):                        0.305\n",
+       "Kurtosis:                       3.548   Cond. No.                         1.08\n",
+       "==============================================================================\n",
+       "\n",
+       "Warnings:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "m_7_1 = smf.ols(\"brain_std ~ mass_std\", data=brains).fit()\n",
+    "m_7_1.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.4"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "arviz.stats.stats_utils - WARNING - Shape validation failed: input_shape: (1, 1000), minimum_shape: (chains=2, draws=4)\n",
+      "arviz.stats.stats_utils - WARNING - Shape validation failed: input_shape: (1, 1000), minimum_shape: (chains=2, draws=4)\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>mean</th>\n",
+       "      <th>sd</th>\n",
+       "      <th>hpd_5.5%</th>\n",
+       "      <th>hpd_94.5%</th>\n",
+       "      <th>mcse_mean</th>\n",
+       "      <th>mcse_sd</th>\n",
+       "      <th>ess_mean</th>\n",
+       "      <th>ess_sd</th>\n",
+       "      <th>ess_bulk</th>\n",
+       "      <th>ess_tail</th>\n",
+       "      <th>r_hat</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>b</th>\n",
+       "      <td>0.169</td>\n",
+       "      <td>0.074</td>\n",
+       "      <td>0.058</td>\n",
+       "      <td>0.294</td>\n",
+       "      <td>0.002</td>\n",
+       "      <td>0.002</td>\n",
+       "      <td>1021.0</td>\n",
+       "      <td>956.0</td>\n",
+       "      <td>1020.0</td>\n",
+       "      <td>944.0</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>a</th>\n",
+       "      <td>0.528</td>\n",
+       "      <td>0.069</td>\n",
+       "      <td>0.410</td>\n",
+       "      <td>0.634</td>\n",
+       "      <td>0.002</td>\n",
+       "      <td>0.002</td>\n",
+       "      <td>867.0</td>\n",
+       "      <td>798.0</td>\n",
+       "      <td>882.0</td>\n",
+       "      <td>983.0</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    mean     sd  hpd_5.5%  hpd_94.5%  mcse_mean  mcse_sd  ess_mean  ess_sd  \\\n",
+       "b  0.169  0.074     0.058      0.294      0.002    0.002    1021.0   956.0   \n",
+       "a  0.528  0.069     0.410      0.634      0.002    0.002     867.0   798.0   \n",
+       "\n",
+       "   ess_bulk  ess_tail  r_hat  \n",
+       "b    1020.0     944.0    NaN  \n",
+       "a     882.0     983.0    NaN  "
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p, cov = np.polyfit(brains.loc[:, \"mass_std\"], brains.loc[:, \"brain_std\"], 1, cov=True)\n",
+    "\n",
+    "post = stats.multivariate_normal(p, cov).rvs(1000)\n",
+    "\n",
+    "az.summary({k: v for k, v in zip(\"ba\", post.T)})"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.5"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.49015804794908413"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "1 - m_7_1.resid.var() / brains.brain_std.var()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.6"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.49015804794908413"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "def R2_is_bad(model):\n",
+    "    return 1 - model.resid.var() / brains.brain_std.var()\n",
+    "\n",
+    "\n",
+    "R2_is_bad(m_7_1)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.7"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/statsmodels/stats/stattools.py:71: ValueWarning: omni_normtest is not valid with less than 8 observations; 7 samples were given.\n",
+      "  \"samples were given.\" % int(n), ValueWarning)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>        <td>brain_std</td>    <th>  R-squared:         </th> <td>   0.536</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.304</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   2.310</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Tue, 19 May 2020</td> <th>  Prob (F-statistic):</th>  <td> 0.215</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>16:36:52</td>     <th>  Log-Likelihood:    </th> <td>  3.3223</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>     7</td>      <th>  AIC:               </th> <td> -0.6445</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>     4</td>      <th>  BIC:               </th> <td> -0.8068</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "          <td></td>            <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>        <td>    0.6142</td> <td>    0.155</td> <td>    3.950</td> <td> 0.017</td> <td>    0.182</td> <td>    1.046</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>mass_std</th>         <td>    0.1957</td> <td>    0.093</td> <td>    2.101</td> <td> 0.104</td> <td>   -0.063</td> <td>    0.454</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>I(mass_std ** 2)</th> <td>   -0.0998</td> <td>    0.159</td> <td>   -0.629</td> <td> 0.564</td> <td>   -0.540</td> <td>    0.341</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>   nan</td> <th>  Durbin-Watson:     </th> <td>   1.479</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td>   nan</td> <th>  Jarque-Bera (JB):  </th> <td>   1.016</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 0.901</td> <th>  Prob(JB):          </th> <td>   0.602</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 2.514</td> <th>  Cond. No.          </th> <td>    4.04</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:              brain_std   R-squared:                       0.536\n",
+       "Model:                            OLS   Adj. R-squared:                  0.304\n",
+       "Method:                 Least Squares   F-statistic:                     2.310\n",
+       "Date:                Tue, 19 May 2020   Prob (F-statistic):              0.215\n",
+       "Time:                        16:36:52   Log-Likelihood:                 3.3223\n",
+       "No. Observations:                   7   AIC:                           -0.6445\n",
+       "Df Residuals:                       4   BIC:                           -0.8068\n",
+       "Df Model:                           2                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "====================================================================================\n",
+       "                       coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------------\n",
+       "Intercept            0.6142      0.155      3.950      0.017       0.182       1.046\n",
+       "mass_std             0.1957      0.093      2.101      0.104      -0.063       0.454\n",
+       "I(mass_std ** 2)    -0.0998      0.159     -0.629      0.564      -0.540       0.341\n",
+       "==============================================================================\n",
+       "Omnibus:                          nan   Durbin-Watson:                   1.479\n",
+       "Prob(Omnibus):                    nan   Jarque-Bera (JB):                1.016\n",
+       "Skew:                           0.901   Prob(JB):                        0.602\n",
+       "Kurtosis:                       2.514   Cond. No.                         4.04\n",
+       "==============================================================================\n",
+       "\n",
+       "Warnings:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "m_7_2 = smf.ols(\"brain_std ~ mass_std + I(mass_std**2)\", data=brains).fit()\n",
+    "m_7_2.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.8"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "m_7_3 = smf.ols(\n",
+    "    \"brain_std ~ mass_std + I(mass_std**2) + I(mass_std**3)\", data=brains\n",
+    ").fit()\n",
+    "m_7_4 = smf.ols(\n",
+    "    \"brain_std ~ mass_std + I(mass_std**2) + I(mass_std**3) + I(mass_std**4)\",\n",
+    "    data=brains,\n",
+    ").fit()\n",
+    "m_7_5 = smf.ols(\n",
+    "    \"brain_std ~ mass_std + I(mass_std**2) + I(mass_std**3) + I(mass_std**4) + I(mass_std**5)\",\n",
+    "    data=brains,\n",
+    ").fit()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.9"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "m_7_6 = smf.ols(\n",
+    "    \"brain_std ~ mass_std + I(mass_std**2) + I(mass_std**3) + I(mass_std**4) + I(mass_std**5) + I(mass_std**6)\",\n",
+    "    data=brains,\n",
+    ").fit()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.10\n",
+    "\n",
+    "The chapter gives code to produce the first panel of Figure 7.3. Here, produce the entire figure by looping over models 7.1-7.6.\n",
+    "\n",
+    "To sample the posterior predictive on a new independent variable we make use of theano SharedVariable objects, as outlined [here](https://docs.pymc.io/notebooks/data_container.html)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x850 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "models = [m_7_1, m_7_2, m_7_3, m_7_4, m_7_5, m_7_6]\n",
+    "names = [\"m_7_1\", \"m_7_2\", \"m_7_3\", \"m_7_4\", \"m_7_5\", \"m_7_6\"]\n",
+    "\n",
+    "mass_plot = np.linspace(33, 62, 100)\n",
+    "mass_new = (mass_plot - brains.mass.mean()) / brains.mass.std()\n",
+    "\n",
+    "fig, axs = plt.subplots(3, 2, figsize=[6, 8.5], sharex=True, sharey=\"row\")\n",
+    "\n",
+    "for model, name, ax in zip(models, names, axs.flat):\n",
+    "    prediction = model.get_prediction({\"mass_std\": mass_new})\n",
+    "    pred = prediction.summary_frame(alpha=0.11) * brains.brain.max()\n",
+    "\n",
+    "    ax.plot(mass_plot, pred[\"mean\"])\n",
+    "    ax.fill_between(mass_plot, pred[\"mean_ci_lower\"], pred[\"mean_ci_upper\"], alpha=0.3)\n",
+    "    ax.scatter(brains.mass, brains.brain, color=\"C0\", s=15)\n",
+    "\n",
+    "    ax.set_title(f\"{name}: R^2: {model.rsquared:.2f}\", loc=\"left\", fontsize=11)\n",
+    "\n",
+    "    if ax.is_first_col():\n",
+    "        ax.set_ylabel(\"brain volume (cc)\")\n",
+    "\n",
+    "    if ax.is_last_row():\n",
+    "        ax.set_xlabel(\"body mass (kg)\")\n",
+    "\n",
+    "    if ax.is_last_row():\n",
+    "        ax.set_ylim(-500, 2100)\n",
+    "        ax.axhline(0, ls=\"dashed\", c=\"k\", lw=1)\n",
+    "        ax.set_yticks([0, 450, 1300])\n",
+    "    else:\n",
+    "        ax.set_ylim(300, 1600)\n",
+    "        ax.set_yticks([450, 900, 1300])\n",
+    "\n",
+    "fig.tight_layout()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.11 - this is R specific notation for dropping rows"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "brains_new = brains.drop(brains.index[-1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 800x300 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Figure 7.4\n",
+    "\n",
+    "# this code taken from PyMC3 port of Rethinking/Chp_06.ipynb\n",
+    "\n",
+    "f, (ax1, ax2) = plt.subplots(1, 2, sharey=True, figsize=(8, 3))\n",
+    "ax1.scatter(brains.mass, brains.brain, alpha=0.8)\n",
+    "ax2.scatter(brains.mass, brains.brain, alpha=0.8)\n",
+    "for i in range(len(brains)):\n",
+    "    d_new = brains.drop(brains.index[-i])  # drop each data point in turn\n",
+    "\n",
+    "    # first order model\n",
+    "    m0 = smf.ols(\"brain ~ mass\", d_new).fit()\n",
+    "    # need to calculate regression line\n",
+    "    # need to add intercept term explicitly\n",
+    "    x = sm.add_constant(d_new.mass)  # add constant to new data frame with mass\n",
+    "    x_pred = pd.DataFrame(\n",
+    "        {\"mass\": np.linspace(x.mass.min() - 10, x.mass.max() + 10, 50)}\n",
+    "    )  # create linspace dataframe\n",
+    "    x_pred2 = sm.add_constant(\n",
+    "        x_pred\n",
+    "    )  # add constant to newly created linspace dataframe\n",
+    "    y_pred = m0.predict(x_pred2)  # calculate predicted values\n",
+    "    ax1.plot(x_pred, y_pred, \"gray\", alpha=0.5)\n",
+    "    ax1.set_ylabel(\"body mass (kg)\", fontsize=12)\n",
+    "    ax1.set_xlabel(\"brain volume (cc)\", fontsize=12)\n",
+    "    ax1.set_title(\"Underfit model\")\n",
+    "\n",
+    "    # fifth order model\n",
+    "    m1 = smf.ols(\n",
+    "        \"brain ~ mass + I(mass**2) + I(mass**3) + I(mass**4) + I(mass**5)\", data=d_new\n",
+    "    ).fit()\n",
+    "    x = sm.add_constant(d_new.mass)  # add constant to new data frame with mass\n",
+    "    x_pred = pd.DataFrame(\n",
+    "        {\"mass\": np.linspace(x.mass.min() - 10, x.mass.max() + 10, 200)}\n",
+    "    )  # create linspace dataframe\n",
+    "    x_pred2 = sm.add_constant(\n",
+    "        x_pred\n",
+    "    )  # add constant to newly created linspace dataframe\n",
+    "    y_pred = m1.predict(x_pred2)  # calculate predicted values from fitted model\n",
+    "    ax2.plot(x_pred, y_pred, \"gray\", alpha=0.5)\n",
+    "    ax2.set_xlim(32, 62)\n",
+    "    ax2.set_ylim(-250, 2200)\n",
+    "    ax2.set_ylabel(\"body mass (kg)\", fontsize=12)\n",
+    "    ax2.set_xlabel(\"brain volume (cc)\", fontsize=12)\n",
+    "    ax2.set_title(\"Overfit model\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.12"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.6108643020548935"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p = np.array([0.3, 0.7])\n",
+    "-np.sum(p * np.log(p))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Figure 7.5\n",
+    "p = np.array([0.3, 0.7])\n",
+    "q = np.arange(0.01, 1, 0.01)\n",
+    "DKL = np.sum(p * np.log(p / np.array([q, 1 - q]).T), 1)\n",
+    "\n",
+    "plt.plot(q, DKL)\n",
+    "plt.xlabel(\"q[1]\")\n",
+    "plt.ylabel(\"Divergence of q from p\")\n",
+    "plt.axvline(0.3, ls=\"dashed\", color=\"k\")\n",
+    "plt.text(0.315, 1.22, \"q = p\");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.13 & 7.14"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "n_samples = 3000\n",
+    "\n",
+    "intercept, slope = (\n",
+    "    stats.multivariate_normal(m_7_1.params, m_7_1.cov_params()).rvs(n_samples).T\n",
+    ")\n",
+    "\n",
+    "pred = intercept + slope * brains.mass_std.values.reshape(-1, 1)\n",
+    "\n",
+    "n, ns = pred.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([ 0.62060273,  0.66080837,  0.56608632,  0.62876608,  0.47909493,\n",
+       "        0.44851465, -0.8559313 ])"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# PyMC3 does not have a way to calculate LPPD directly, so we use the approach from 7.14\n",
+    "\n",
+    "sigmas = (np.sum((pred - brains.brain_std.values.reshape(-1, 1)) ** 2, 0) / 7) ** 0.5\n",
+    "ll = np.zeros((n, ns))\n",
+    "for s in range(ns):\n",
+    "    logprob = stats.norm.logpdf(brains.brain_std, pred[:, s], sigmas[s])\n",
+    "    ll[:, s] = logprob\n",
+    "\n",
+    "lppd = np.zeros(n)\n",
+    "for i in range(n):\n",
+    "    lppd[i] = logsumexp(ll[i]) - np.log(ns)\n",
+    "\n",
+    "lppd"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.15"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# make an lppd function that can be applied to all models (from code above)\n",
+    "def lppd(model, n_samples=1e4):\n",
+    "    n_samples = int(n_samples)\n",
+    "\n",
+    "    pars = stats.multivariate_normal(model.params, model.cov_params()).rvs(n_samples).T\n",
+    "    dmat = dmatrix(\n",
+    "        model.model.data.design_info, brains, return_type=\"dataframe\"\n",
+    "    ).values  # get model design matrix\n",
+    "    pred = dmat.dot(pars)\n",
+    "\n",
+    "    n, ns = pred.shape\n",
+    "\n",
+    "    # this approach for calculating lppd isfrom 7.14\n",
+    "    sigmas = (\n",
+    "        np.sum((pred - brains.brain_std.values.reshape(-1, 1)) ** 2, 0) / 7\n",
+    "    ) ** 0.5\n",
+    "    ll = np.zeros((n, ns))\n",
+    "    for s in range(ns):\n",
+    "        logprob = stats.norm.logpdf(brains.brain_std, pred[:, s], sigmas[s])\n",
+    "        ll[:, s] = logprob\n",
+    "\n",
+    "    lppd = np.zeros(n)\n",
+    "    for i in range(n):\n",
+    "        lppd[i] = logsumexp(ll[i]) - np.log(ns)\n",
+    "\n",
+    "    return lppd"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([ 2.54799078,  2.3115561 ,  2.89843278,  3.53621752, 11.04367575])"
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# model 7_6 does not work with OLS because its covariance matrix is not finite.\n",
+    "lppds = np.array(list(map(lppd, models[:-1], [1000] * len(models[:-1]))))\n",
+    "\n",
+    "lppds.sum(1)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.16\n",
+    "\n",
+    "This relies on the `sim.train.test` function in the `rethinking` package. [This](https://github.com/rmcelreath/rethinking/blob/master/R/sim_train_test.R) is the original function.\n",
+    "\n",
+    "The python port of this function below is from [Rethinking/Chp_06](https://nbviewer.jupyter.org/github/pymc-devs/resources/blob/master/Rethinking/Chp_06.ipynb) Code 6.12."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def sim_train_test(N=20, k=3, rho=[0.15, -0.4], b_sigma=100):\n",
+    "\n",
+    "    n_dim = 1 + len(rho)\n",
+    "    if n_dim < k:\n",
+    "        n_dim = k\n",
+    "    Rho = np.diag(np.ones(n_dim))\n",
+    "    Rho[0, 1:3:1] = rho\n",
+    "    i_lower = np.tril_indices(n_dim, -1)\n",
+    "    Rho[i_lower] = Rho.T[i_lower]\n",
+    "\n",
+    "    x_train = stats.multivariate_normal.rvs(cov=Rho, size=N)\n",
+    "    x_test = stats.multivariate_normal.rvs(cov=Rho, size=N)\n",
+    "\n",
+    "    mm_train = np.ones((N, 1))\n",
+    "\n",
+    "    np.concatenate([mm_train, x_train[:, 1:k]], axis=1)\n",
+    "\n",
+    "    # Using pymc3\n",
+    "\n",
+    "    with pm.Model() as m_sim:\n",
+    "        vec_V = pm.MvNormal(\n",
+    "            \"vec_V\",\n",
+    "            mu=0,\n",
+    "            cov=b_sigma * np.eye(n_dim),\n",
+    "            shape=(1, n_dim),\n",
+    "            testval=np.random.randn(1, n_dim) * 0.01,\n",
+    "        )\n",
+    "        mu = pm.Deterministic(\"mu\", 0 + pm.math.dot(x_train, vec_V.T))\n",
+    "        y = pm.Normal(\"y\", mu=mu, sd=1, observed=x_train[:, 0])\n",
+    "\n",
+    "    with m_sim:\n",
+    "        trace_m_sim = pm.sample()\n",
+    "\n",
+    "    vec = pm.summary(trace_m_sim)[\"mean\"][:n_dim]\n",
+    "    vec = np.array([i for i in vec]).reshape(n_dim, -1)\n",
+    "\n",
+    "    dev_train = -2 * sum(\n",
+    "        stats.norm.logpdf(x_train, loc=np.matmul(x_train, vec), scale=1)\n",
+    "    )\n",
+    "\n",
+    "    mm_test = np.ones((N, 1))\n",
+    "\n",
+    "    mm_test = np.concatenate([mm_test, x_test[:, 1 : k + 1]], axis=1)\n",
+    "\n",
+    "    dev_test = -2 * sum(\n",
+    "        stats.norm.logpdf(x_test[:, 0], loc=np.matmul(mm_test, vec), scale=1)\n",
+    "    )\n",
+    "\n",
+    "    return np.mean(dev_train), np.mean(dev_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 986.17draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 940.71draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 805.99draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1148.87draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 732.77draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1132.31draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1097.05draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 856.80draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1117.58draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 968.67draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "3\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 952.71draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 845.98draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 859.55draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1025.12draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1050.64draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1016.58draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 949.01draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 948.06draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1143.00draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 876.46draws/s] \n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "4\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1104.35draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1053.62draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1018.15draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1025.73draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1063.74draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1153.38draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 983.81draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 866.76draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1122.08draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1047.47draws/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "5\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 870.17draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 979.60draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1073.69draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1091.23draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1122.36draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1016.52draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 955.24draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1030.75draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1087.71draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1025.17draws/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "6\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 962.10draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1044.93draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:02<00:00, 893.47draws/s] \n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1080.25draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1123.38draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1002.28draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1136.63draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1064.63draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1045.91draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [vec_V]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1141.00draws/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "n = 20\n",
+    "tries = 10\n",
+    "param = 6\n",
+    "r = np.zeros(shape=(param - 1, 4))\n",
+    "\n",
+    "train = []\n",
+    "test = []\n",
+    "\n",
+    "for j in range(2, param + 1):\n",
+    "    print(j)\n",
+    "    for i in range(1, tries + 1):\n",
+    "        tr, te = sim_train_test(N=n, k=param)\n",
+    "        train.append(tr), test.append(te)\n",
+    "    r[j - 2, :] = (\n",
+    "        np.mean(train),\n",
+    "        np.std(train, ddof=1),\n",
+    "        np.mean(test),\n",
+    "        np.std(test, ddof=1),\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.17\n",
+    "\n",
+    "Does not apply because multi-threading is automatic in PyMC3."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.18"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "num_param = np.arange(2, param + 1)\n",
+    "\n",
+    "plt.figure(figsize=(10, 6))\n",
+    "plt.scatter(num_param, r[:, 0], color=\"C0\")\n",
+    "plt.xticks(num_param)\n",
+    "\n",
+    "for j in range(param - 1):\n",
+    "    plt.vlines(\n",
+    "        num_param[j],\n",
+    "        r[j, 0] - r[j, 1],\n",
+    "        r[j, 0] + r[j, 1],\n",
+    "        color=\"mediumblue\",\n",
+    "        zorder=-1,\n",
+    "        alpha=0.80,\n",
+    "    )\n",
+    "\n",
+    "plt.scatter(num_param + 0.1, r[:, 2], facecolors=\"none\", edgecolors=\"k\")\n",
+    "\n",
+    "for j in range(param - 1):\n",
+    "    plt.vlines(\n",
+    "        num_param[j] + 0.1,\n",
+    "        r[j, 2] - r[j, 3],\n",
+    "        r[j, 2] + r[j, 3],\n",
+    "        color=\"k\",\n",
+    "        zorder=-2,\n",
+    "        alpha=0.70,\n",
+    "    )\n",
+    "\n",
+    "dist = 0.20\n",
+    "plt.text(num_param[1] - dist, r[1, 0] - dist, \"in\", color=\"C0\", fontsize=13)\n",
+    "plt.text(num_param[1] + dist, r[1, 2] - dist, \"out\", color=\"k\", fontsize=13)\n",
+    "plt.text(num_param[1] + dist, r[1, 2] + r[1, 3] - dist, \"+1 SD\", color=\"k\", fontsize=10)\n",
+    "plt.text(num_param[1] + dist, r[1, 2] - r[1, 3] - dist, \"+1 SD\", color=\"k\", fontsize=10)\n",
+    "plt.xlabel(\"Number of parameters\", fontsize=14)\n",
+    "plt.ylabel(\"Deviance\", fontsize=14)\n",
+    "plt.title(\"N = {}\".format(n), fontsize=14)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "These uncertainties are a *lot* larger than in the book... MCMC vs OLS again?"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.19\n",
+    "\n",
+    "7.19 to 7.25 transcribed directly from 6.15-6.20 in [Chapter 6 of 1st Edition](https://nbviewer.jupyter.org/github/pymc-devs/resources/blob/master/Rethinking/Chp_06.ipynb)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.read_csv(\"Data/cars.csv\", sep=\",\", index_col=0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, b, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 30000/30000 [00:27<00:00, 1098.57draws/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "with pm.Model() as m:\n",
+    "    a = pm.Normal(\"a\", mu=0, sd=100)\n",
+    "    b = pm.Normal(\"b\", mu=0, sd=10)\n",
+    "    sigma = pm.Uniform(\"sigma\", 0, 30)\n",
+    "    mu = pm.Deterministic(\"mu\", a + b * data[\"speed\"])\n",
+    "    dist = pm.Normal(\"dist\", mu=mu, sd=sigma, observed=data[\"dist\"])\n",
+    "    m = pm.sample(5000, tune=10000)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.20"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "n_samples = 1000\n",
+    "n_cases = data.shape[0]\n",
+    "logprob = np.zeros((n_cases, n_samples))\n",
+    "\n",
+    "for s in range(0, n_samples):\n",
+    "    mu = m[\"a\"][s] + m[\"b\"][s] * data[\"speed\"]\n",
+    "    p_ = stats.norm.logpdf(data[\"dist\"], loc=mu, scale=m[\"sigma\"][s])\n",
+    "    logprob[:, s] = p_"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.21"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "n_cases = data.shape[0]\n",
+    "lppd = np.zeros((n_cases))\n",
+    "for a in range(1, n_cases):\n",
+    "    lppd[a,] = logsumexp(logprob[a,]) - np.log(n_samples)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.22"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pWAIC = np.zeros((n_cases))\n",
+    "for i in range(1, n_cases):\n",
+    "    pWAIC[i,] = np.var(logprob[i,])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.23"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "412.2473721341997"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "-2 * (sum(lppd) - sum(pWAIC))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.24"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "15.062775341988091"
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "waic_vec = -2 * (lppd - pWAIC)\n",
+    "(n_cases * np.var(waic_vec)) ** 0.5"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Setup for Code 7.25+\n",
+    "\n",
+    "Have to reproduce m6.6-m6.8 from Code 6.13-6.17 in Chapter 6"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, p]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1642.13draws/s]\n",
+      "The acceptance probability does not match the target. It is 0.8790785610439991, but should be close to 0.8. Try to increase the number of tuning steps.\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, bf, bt, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1014.87draws/s]\n",
+      "The acceptance probability does not match the target. It is 0.8878556060097744, but should be close to 0.8. Try to increase the number of tuning steps.\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, bt, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1342.39draws/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "# number of plants\n",
+    "N = 100\n",
+    "# simulate initial heights\n",
+    "h0 = np.random.normal(10, 2, N)\n",
+    "# assign treatments and simulate fungus and growth\n",
+    "treatment = np.repeat([0, 1], N / 2)\n",
+    "fungus = np.random.binomial(n=1, p=0.5 - treatment * 0.4, size=N)\n",
+    "h1 = h0 + np.random.normal(5 - 3 * fungus, size=N)\n",
+    "# compose a clean data frame\n",
+    "d = pd.DataFrame.from_dict(\n",
+    "    {\"h0\": h0, \"h1\": h1, \"treatment\": treatment, \"fungus\": fungus}\n",
+    ")\n",
+    "\n",
+    "with pm.Model() as m_6_6:\n",
+    "    p = pm.Lognormal(\"p\", 0, 0.25)\n",
+    "\n",
+    "    mu = pm.Deterministic(\"mu\", p * d.h0)\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    h1 = pm.Normal(\"h1\", mu=mu, sigma=sigma, observed=d.h1)\n",
+    "\n",
+    "    m_6_6_trace = pm.sample()\n",
+    "\n",
+    "with pm.Model() as m_6_7:\n",
+    "    a = pm.Normal(\"a\", 0, 0.2)\n",
+    "    bt = pm.Normal(\"bt\", 0, 0.5)\n",
+    "    bf = pm.Normal(\"bf\", 0, 0.5)\n",
+    "\n",
+    "    p = a + bt * d.treatment + bf * d.fungus\n",
+    "\n",
+    "    mu = pm.Deterministic(\"mu\", p * d.h0)\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    h1 = pm.Normal(\"h1\", mu=mu, sigma=sigma, observed=d.h1)\n",
+    "\n",
+    "    m_6_7_trace = pm.sample()\n",
+    "\n",
+    "with pm.Model() as m_6_8:\n",
+    "    a = pm.Normal(\"a\", 0, 0.2)\n",
+    "    bt = pm.Normal(\"bt\", 0, 0.5)\n",
+    "\n",
+    "    p = a + bt * d.treatment\n",
+    "\n",
+    "    mu = pm.Deterministic(\"mu\", p * d.h0)\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    h1 = pm.Normal(\"h1\", mu=mu, sigma=sigma, observed=d.h1)\n",
+    "\n",
+    "    m_6_8_trace = pm.sample()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.25"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/arviz/stats/stats.py:1210: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
+      "See http://arxiv.org/abs/1507.04544 for details\n",
+      "  \"For one or more samples the posterior variance of the log predictive \"\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "Computed from 1000 by 100 log-likelihood matrix\n",
+       "\n",
+       "              Estimate       SE\n",
+       "deviance_waic   361.57    15.09\n",
+       "p_waic            3.32        -\n",
+       "\n",
+       "There has been a warning during the calculation. Please check the results."
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "az.waic(m_6_7_trace, m_6_7, scale=\"deviance\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.26"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "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>rank</th>\n",
+       "      <th>waic</th>\n",
+       "      <th>p_waic</th>\n",
+       "      <th>d_waic</th>\n",
+       "      <th>weight</th>\n",
+       "      <th>se</th>\n",
+       "      <th>dse</th>\n",
+       "      <th>warning</th>\n",
+       "      <th>waic_scale</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>m_6_7</th>\n",
+       "      <td>0</td>\n",
+       "      <td>361.565</td>\n",
+       "      <td>3.31507</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>15.0903</td>\n",
+       "      <td>0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>deviance</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_6_8</th>\n",
+       "      <td>1</td>\n",
+       "      <td>410.045</td>\n",
+       "      <td>2.41389</td>\n",
+       "      <td>48.48</td>\n",
+       "      <td>2.9697e-11</td>\n",
+       "      <td>11.3292</td>\n",
+       "      <td>9.96148</td>\n",
+       "      <td>False</td>\n",
+       "      <td>deviance</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_6_6</th>\n",
+       "      <td>2</td>\n",
+       "      <td>418.332</td>\n",
+       "      <td>1.57451</td>\n",
+       "      <td>56.7666</td>\n",
+       "      <td>4.71286e-13</td>\n",
+       "      <td>11.7292</td>\n",
+       "      <td>11.8463</td>\n",
+       "      <td>False</td>\n",
+       "      <td>deviance</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      rank     waic   p_waic   d_waic       weight       se      dse warning  \\\n",
+       "m_6_7    0  361.565  3.31507        0            1  15.0903        0    True   \n",
+       "m_6_8    1  410.045  2.41389    48.48   2.9697e-11  11.3292  9.96148   False   \n",
+       "m_6_6    2  418.332  1.57451  56.7666  4.71286e-13  11.7292  11.8463   False   \n",
+       "\n",
+       "      waic_scale  \n",
+       "m_6_7   deviance  \n",
+       "m_6_8   deviance  \n",
+       "m_6_6   deviance  "
+      ]
+     },
+     "execution_count": 34,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "compare_df = az.compare(\n",
+    "    {\"m_6_6\": m_6_6_trace, \"m_6_7\": m_6_7_trace, \"m_6_8\": m_6_8_trace,},\n",
+    "    method=\"pseudo-BMA\",\n",
+    "    ic=\"waic\",\n",
+    "    scale=\"deviance\",\n",
+    ")\n",
+    "compare_df"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.27"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/arviz/stats/stats.py:1210: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
+      "See http://arxiv.org/abs/1507.04544 for details\n",
+      "  \"For one or more samples the posterior variance of the log predictive \"\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "array(9.96147766)"
+      ]
+     },
+     "execution_count": 35,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "waic_m_6_7 = az.waic(m_6_7_trace, pointwise=True, scale=\"deviance\")\n",
+    "waic_m_6_8 = az.waic(m_6_8_trace, pointwise=True, scale=\"deviance\")\n",
+    "\n",
+    "# pointwise values are stored in the waic_i attribute.\n",
+    "diff_m_6_7_m_6_8 = waic_m_6_7.waic_i - waic_m_6_8.waic_i\n",
+    "\n",
+    "n = len(diff_m_6_7_m_6_8)\n",
+    "\n",
+    "np.sqrt(n * np.var(diff_m_6_7_m_6_8)).values"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.28"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([12.96, 67.04])"
+      ]
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "40.0 + np.array([-1, 1]) * 10.4 * 2.6"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.29"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x300 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "az.plot_compare(compare_df);"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.30"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array(0.69300909)"
+      ]
+     },
+     "execution_count": 38,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "waic_m_6_6 = az.waic(m_6_6_trace, pointwise=True, scale=\"deviance\")\n",
+    "\n",
+    "diff_m6_6_m6_8 = waic_m_6_6.waic_i - waic_m_6_8.waic_i\n",
+    "\n",
+    "np.sqrt(1 * np.var(diff_m6_6_m6_8)).values"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.31\n",
+    "\n",
+    "dSE is calculated by compare above, but `rethinking` produces a pairwise comparison. This is not implemented in `arviz`, but we can hack it together:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/arviz/stats/stats.py:1210: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
+      "See http://arxiv.org/abs/1507.04544 for details\n",
+      "  \"For one or more samples the posterior variance of the log predictive \"\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>m_6_7</th>\n",
+       "      <th>m_6_8</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>m_6_8</th>\n",
+       "      <td>9.96148</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_6_6</th>\n",
+       "      <td>11.8463</td>\n",
+       "      <td>6.93009</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         m_6_7    m_6_8\n",
+       "m_6_8  9.96148        0\n",
+       "m_6_6  11.8463  6.93009"
+      ]
+     },
+     "execution_count": 39,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset_dict = {\"m_6_6\": m_6_6_trace, \"m_6_7\": m_6_7_trace, \"m_6_8\": m_6_8_trace}\n",
+    "\n",
+    "# compare all models\n",
+    "s0 = az.compare(dataset_dict, ic=\"waic\", scale=\"deviance\")[\"dse\"]\n",
+    "# the output compares each model to the 'best' model - i.e. two models are compared to one.\n",
+    "# to complete a pair-wise comparison we need to compare the remaining two models.\n",
+    "\n",
+    "# to do this, remove the 'best' model from the input data\n",
+    "del dataset_dict[s0.index[0]]\n",
+    "\n",
+    "# re-run compare with the remaining two models\n",
+    "s1 = az.compare(dataset_dict, ic=\"waic\", scale=\"deviance\")[\"dse\"]\n",
+    "\n",
+    "# s0 compares two models to one model, and s1 compares the remaining two models to each other\n",
+    "# now we just nee to wrangle them together!\n",
+    "\n",
+    "# convert them both to dataframes, setting the name to the 'best' model in each `compare` output.\n",
+    "# (i.e. the name is the model that others are compared to)\n",
+    "df_0 = s0.to_frame(name=s0.index[0])\n",
+    "df_1 = s1.to_frame(name=s1.index[0])\n",
+    "\n",
+    "# merge these dataframes to create a pairwise comparison\n",
+    "pd.merge(df_0, df_1, left_index=True, right_index=True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**Note:** this work for three models, but will get increasingly hack-y with additional models. The function below can be applied to *n* models:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def pairwise_compare(dataset_dict, metric=\"dse\", **kwargs):\n",
+    "    \"\"\"\n",
+    "    Calculate pairwise comparison of models in dataset_dict.\n",
+    "    \n",
+    "    Parameters\n",
+    "    ----------\n",
+    "    dataset_dict : dict\n",
+    "        A dict containing two ore more {'name': pymc3.backends.base.MultiTrace}\n",
+    "        items.\n",
+    "    metric : str\n",
+    "        The name of the matric to be calculated. Can be any valid column output\n",
+    "        by `arviz.compare`. Note that this may change depending on the **kwargs\n",
+    "        that are specified.\n",
+    "    kwargs\n",
+    "        Arguments passed to `arviz.compare`\n",
+    "    \"\"\"\n",
+    "    data_dict = dataset_dict.copy()\n",
+    "    dicts = []\n",
+    "\n",
+    "    while len(data_dict) > 1:\n",
+    "        c = az.compare(data_dict, **kwargs)[metric]\n",
+    "        dicts.append(c.to_frame(name=c.index[0]))\n",
+    "        del data_dict[c.index[0]]\n",
+    "\n",
+    "    return pd.concat(dicts, axis=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/arviz/stats/stats.py:1210: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
+      "See http://arxiv.org/abs/1507.04544 for details\n",
+      "  \"For one or more samples the posterior variance of the log predictive \"\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>m_6_7</th>\n",
+       "      <th>m_6_8</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>m_6_7</th>\n",
+       "      <td>0</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_6_8</th>\n",
+       "      <td>9.96148</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_6_6</th>\n",
+       "      <td>11.8463</td>\n",
+       "      <td>6.93009</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         m_6_7    m_6_8\n",
+       "m_6_7        0      NaN\n",
+       "m_6_8  9.96148        0\n",
+       "m_6_6  11.8463  6.93009"
+      ]
+     },
+     "execution_count": 41,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset_dict = {\"m_6_6\": m_6_6_trace, \"m_6_7\": m_6_7_trace, \"m_6_8\": m_6_8_trace}\n",
+    "\n",
+    "pairwise_compare(dataset_dict, metric=\"dse\", ic=\"waic\", scale=\"deviance\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.32"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "d = pd.read_csv(\"Data/WaffleDivorce.csv\", delimiter=\";\")\n",
+    "\n",
+    "d[\"A\"] = stats.zscore(d[\"MedianAgeMarriage\"])\n",
+    "d[\"D\"] = stats.zscore(d[\"Divorce\"])\n",
+    "d[\"M\"] = stats.zscore(d[\"Marriage\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, bA, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1660.43draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, bM, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1952.88draws/s]\n",
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, bM, bA, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1450.27draws/s]\n",
+      "The acceptance probability does not match the target. It is 0.884650266255495, but should be close to 0.8. Try to increase the number of tuning steps.\n"
+     ]
+    }
+   ],
+   "source": [
+    "with pm.Model() as m_5_1:\n",
+    "    a = pm.Normal(\"a\", 0, 0.2)\n",
+    "    bA = pm.Normal(\"bA\", 0, 0.5)\n",
+    "\n",
+    "    mu = a + bA * d[\"A\"]\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    D = pm.Normal(\"D\", mu, sigma, observed=d[\"D\"])\n",
+    "\n",
+    "    m_5_1_trace = pm.sample()\n",
+    "\n",
+    "with pm.Model() as m_5_2:\n",
+    "    a = pm.Normal(\"a\", 0, 0.2)\n",
+    "    bM = pm.Normal(\"bM\", 0, 0.5)\n",
+    "\n",
+    "    mu = a + bM * d[\"M\"]\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    D = pm.Normal(\"D\", mu, sigma, observed=d[\"D\"])\n",
+    "\n",
+    "    m_5_2_trace = pm.sample()\n",
+    "\n",
+    "with pm.Model() as m_5_3:\n",
+    "    a = pm.Normal(\"a\", 0, 0.2)\n",
+    "    bA = pm.Normal(\"bA\", 0, 0.5)\n",
+    "    bM = pm.Normal(\"bM\", 0, 0.5)\n",
+    "\n",
+    "    mu = a + bA * d[\"A\"] + bM * d[\"M\"]\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    D = pm.Normal(\"D\", mu, sigma, observed=d[\"D\"])\n",
+    "\n",
+    "    m_5_3_trace = pm.sample()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.33"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/arviz/stats/stats.py:532: UserWarning: Estimated shape parameter of Pareto distribution is greater than 0.7 for one or more samples. You should consider using a more robust model, this is because importance sampling is less likely to work well if the marginal posterior and LOO posterior are very different. This is more likely to happen with a non-robust model and highly influential observations.\n",
+      "  \"Estimated shape parameter of Pareto distribution is greater than 0.7 for \"\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>rank</th>\n",
+       "      <th>loo</th>\n",
+       "      <th>p_loo</th>\n",
+       "      <th>d_loo</th>\n",
+       "      <th>weight</th>\n",
+       "      <th>se</th>\n",
+       "      <th>dse</th>\n",
+       "      <th>warning</th>\n",
+       "      <th>loo_scale</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>m_5_1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>126.508</td>\n",
+       "      <td>3.41868</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.691599</td>\n",
+       "      <td>12.3723</td>\n",
+       "      <td>0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>deviance</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_5_3</th>\n",
+       "      <td>1</td>\n",
+       "      <td>128.869</td>\n",
+       "      <td>4.88707</td>\n",
+       "      <td>2.36092</td>\n",
+       "      <td>0.235037</td>\n",
+       "      <td>10.1269</td>\n",
+       "      <td>0.875712</td>\n",
+       "      <td>False</td>\n",
+       "      <td>deviance</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>m_5_2</th>\n",
+       "      <td>2</td>\n",
+       "      <td>140.334</td>\n",
+       "      <td>2.97358</td>\n",
+       "      <td>13.8261</td>\n",
+       "      <td>0.0733639</td>\n",
+       "      <td>12.8529</td>\n",
+       "      <td>8.82372</td>\n",
+       "      <td>False</td>\n",
+       "      <td>deviance</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      rank      loo    p_loo    d_loo     weight       se       dse warning  \\\n",
+       "m_5_1    0  126.508  3.41868        0   0.691599  12.3723         0    True   \n",
+       "m_5_3    1  128.869  4.88707  2.36092   0.235037  10.1269  0.875712   False   \n",
+       "m_5_2    2  140.334  2.97358  13.8261  0.0733639  12.8529   8.82372   False   \n",
+       "\n",
+       "      loo_scale  \n",
+       "m_5_1  deviance  \n",
+       "m_5_3  deviance  \n",
+       "m_5_2  deviance  "
+      ]
+     },
+     "execution_count": 44,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "az.compare(\n",
+    "    {\"m_5_1\": m_5_1_trace, \"m_5_2\": m_5_2_trace, \"m_5_3\": m_5_3_trace}, scale=\"deviance\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.34"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/oscar/miniconda3/envs/py3/lib/python3.7/site-packages/arviz/stats/stats.py:1210: UserWarning: For one or more samples the posterior variance of the log predictive densities exceeds 0.4. This could be indication of WAIC starting to fail. \n",
+      "See http://arxiv.org/abs/1507.04544 for details\n",
+      "  \"For one or more samples the posterior variance of the log predictive \"\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "psis_m_5_3 = az.loo(m_5_3_trace, pointwise=True, scale=\"deviance\")\n",
+    "waic_m_5_3 = az.waic(m_5_3_trace, pointwise=True, scale=\"deviance\")\n",
+    "\n",
+    "# Figure 7.10\n",
+    "plt.scatter(psis_m_5_3.pareto_k, waic_m_5_3.waic_i)\n",
+    "plt.xlabel(\"PSIS Pareto k\")\n",
+    "plt.ylabel(\"WAIC\");"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 800x350 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Figure 7.11\n",
+    "\n",
+    "v = np.linspace(-4, 4, 100)\n",
+    "\n",
+    "g = stats.norm(loc=0, scale=1)\n",
+    "t = stats.t(df=2, loc=0, scale=1)\n",
+    "\n",
+    "fig, (ax, lax) = plt.subplots(1, 2, figsize=[8, 3.5])\n",
+    "\n",
+    "ax.plot(v, g.pdf(v), color=\"b\")\n",
+    "ax.plot(v, t.pdf(v), color=\"k\")\n",
+    "\n",
+    "lax.plot(v, -g.logpdf(v), color=\"b\")\n",
+    "lax.plot(v, -t.logpdf(v), color=\"k\");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Code 7.35"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Auto-assigning NUTS sampler...\n",
+      "Initializing NUTS using jitter+adapt_diag...\n",
+      "Multiprocess sampling (2 chains in 2 jobs)\n",
+      "NUTS: [sigma, bM, bA, a]\n",
+      "Sampling 2 chains, 0 divergences: 100%|██████████| 2000/2000 [00:01<00:00, 1100.60draws/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "with pm.Model() as m_5_3t:\n",
+    "    a = pm.Normal(\"a\", 0, 0.2)\n",
+    "    bA = pm.Normal(\"bA\", 0, 0.5)\n",
+    "    bM = pm.Normal(\"bM\", 0, 0.5)\n",
+    "\n",
+    "    mu = a + bA * d[\"A\"] + bM * d[\"M\"]\n",
+    "    sigma = pm.Exponential(\"sigma\", 1)\n",
+    "\n",
+    "    D = pm.StudentT(\"D\", 2, mu, sigma, observed=d[\"D\"])\n",
+    "\n",
+    "    m_5_3t_trace = pm.sample()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Computed from 1000 by 50 log-likelihood matrix\n",
+       "\n",
+       "             Estimate       SE\n",
+       "deviance_loo   133.53    10.67\n",
+       "p_loo            5.52        -\n",
+       "------\n",
+       "\n",
+       "Pareto k diagnostic values:\n",
+       "                         Count   Pct.\n",
+       "(-Inf, 0.5]   (good)       50  100.0%\n",
+       " (0.5, 0.7]   (ok)          0    0.0%\n",
+       "   (0.7, 1]   (bad)         0    0.0%\n",
+       "   (1, Inf)   (very bad)    0    0.0%"
+      ]
+     },
+     "execution_count": 48,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "az.loo(m_5_3t_trace, pointwise=True, scale=\"deviance\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x350 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "az.plot_forest(\n",
+    "    [m_5_3_trace, m_5_3t_trace], model_names=[\"m_5_3\", \"m_5_3t\"], figsize=[6, 3.5]\n",
+    ");"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "statsmodels.api 0.11.1\n",
+      "arviz           0.7.0\n",
+      "numpy           1.18.1\n",
+      "pymc3           3.8\n",
+      "pandas          1.0.3\n",
+      "last updated: Tue May 19 2020 \n",
+      "\n",
+      "CPython 3.7.6\n",
+      "IPython 7.14.0\n",
+      "watermark 2.0.2\n"
+     ]
+    }
+   ],
+   "source": [
+    "%load_ext watermark\n",
+    "%watermark -n -u -v -iv -w"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Rethinking_2/Data/Primates301.csv b/Rethinking_2/Data/Primates301.csv
new file mode 100644
index 0000000..6a3168f
--- /dev/null
+++ b/Rethinking_2/Data/Primates301.csv
@@ -0,0 +1,302 @@
+"name";"genus";"species";"subspecies";"spp_id";"genus_id";"social_learning";"research_effort";"brain";"body";"group_size";"gestation";"weaning";"longevity";"sex_maturity";"maternal_investment"
+"Allenopithecus_nigroviridis";"Allenopithecus";"nigroviridis";NA;1;1;0;6;58.02;4655;40;NA;106.15;276;NA;NA
+"Allocebus_trichotis";"Allocebus";"trichotis";NA;2;2;0;6;NA;78.09;1;NA;NA;NA;NA;NA
+"Alouatta_belzebul";"Alouatta";"belzebul";NA;3;3;0;15;52.84;6395;7.4;NA;NA;NA;NA;NA
+"Alouatta_caraya";"Alouatta";"caraya";NA;4;3;0;45;52.63;5383;8.9;185.92;323.16;243.6;1276.72;509.08
+"Alouatta_guariba";"Alouatta";"guariba";NA;5;3;0;37;51.7;5175;7.4;NA;NA;NA;NA;NA
+"Alouatta_palliata";"Alouatta";"palliata";NA;6;3;3;79;49.88;6250;13.1;185.42;495.6;300;1578.42;681.02
+"Alouatta_pigra";"Alouatta";"pigra";NA;7;3;0;25;51.13;8915;5.5;185.92;NA;240;NA;NA
+"Alouatta_sara";"Alouatta";"sara";NA;8;3;0;4;59.08;6611.04;NA;NA;NA;NA;NA;NA
+"Alouatta_seniculus";"Alouatta";"seniculus";NA;9;3;0;82;55.22;5950;7.9;189.9;370.04;300;1690.22;559.94
+"Aotus_azarai";"Aotus";"azarai";NA;10;4;0;22;20.67;1205;4.1;NA;229.69;NA;NA;NA
+"Aotus_azarai_boliviensis";"Aotus";"azarai";"boliviensis";11;4;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Aotus_brumbacki";"Aotus";"brumbacki";NA;12;4;0;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Aotus_infulatus";"Aotus";"infulatus";NA;13;4;0;6;NA;NA;NA;NA;NA;NA;NA;NA
+"Aotus_lemurinus";"Aotus";"lemurinus";NA;14;4;0;16;16.3;734;NA;132.23;74.57;216;755.15;206.8
+"Aotus_lemurinus_griseimembra";"Aotus";"lemurinus";"griseimembra";15;4;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Aotus_nancymaae";"Aotus";"nancymaae";NA;16;4;0;5;NA;791.03;4;NA;NA;NA;NA;NA
+"Aotus_nigriceps";"Aotus";"nigriceps";NA;17;4;0;1;NA;958;3.3;NA;NA;NA;NA;NA
+"Aotus_trivirgatus";"Aotus";"trivirgatus";NA;18;4;0;58;16.85;989;3.15;133.47;76.21;303.6;736.6;209.68
+"Aotus_vociferans";"Aotus";"vociferans";NA;19;4;0;12;NA;703;3.3;NA;NA;NA;NA;NA
+"Archaeolemur_majori";"Archaeolemur";"majori";NA;20;5;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Arctocebus_aureus";"Arctocebus";"aureus";NA;21;6;NA;NA;5.88;210;NA;NA;NA;NA;NA;NA
+"Arctocebus_calabarensis";"Arctocebus";"calabarensis";NA;22;6;0;1;6.92;309;1;133.74;109.26;156;298.91;243
+"Ateles_belzebuth";"Ateles";"belzebuth";NA;23;7;0;12;117.02;8167;14.5;138.2;NA;336;NA;NA
+"Ateles_fusciceps";"Ateles";"fusciceps";NA;24;7;0;4;114.24;9025;NA;224.7;482.7;288;1799.68;707.4
+"Ateles_geoffroyi";"Ateles";"geoffroyi";NA;25;7;2;58;105.09;7535;42;226.37;816.35;327.6;2104.57;1042.72
+"Ateles_paniscus";"Ateles";"paniscus";NA;26;7;0;30;103.85;8280;20;228.18;805.41;453.6;2104.57;1033.59
+"Avahi_cleesei";"Avahi";"cleesei";NA;27;8;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Avahi_laniger";"Avahi";"laniger";NA;28;8;0;10;9.86;1207;2;136.15;149.15;NA;NA;285.3
+"Avahi_occidentalis";"Avahi";"occidentalis";NA;29;8;0;6;7.95;801;3;NA;NA;NA;NA;NA
+"Avahi_unicolor";"Avahi";"unicolor";NA;30;8;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Brachyteles_arachnoides";"Brachyteles";"arachnoides";NA;31;9;0;57;NA;10537.31;19.6;221.75;734.82;NA;2876.24;956.57
+"Bunopithecus_hoolock";"Bunopithecus";"hoolock";NA;32;10;0;24;110.68;6728;3.2;232.5;635.13;NA;2689.08;867.63
+"Cacajao_calvus";"Cacajao";"calvus";NA;33;11;0;11;76;3165;23.7;180;339.29;324;1262.74;519.29
+"Cacajao_melanocephalus";"Cacajao";"melanocephalus";NA;34;11;0;8;68.77;2935;30;NA;NA;216;NA;NA
+"Callicebus_donacophilus";"Callicebus";"donacophilus";NA;35;12;0;1;NA;897.67;1;NA;NA;NA;NA;NA
+"Callicebus_hoffmannsi";"Callicebus";"hoffmannsi";NA;36;12;0;NA;NA;1067.61;1;NA;NA;NA;NA;NA
+"Callicebus_moloch";"Callicebus";"moloch";NA;37;12;0;18;NA;958.13;2.95;164;58.85;303.6;1262.74;222.85
+"Callicebus_personatus";"Callicebus";"personatus";NA;38;12;0;19;NA;1390.8;2.35;NA;NA;NA;NA;NA
+"Callicebus_torquatus";"Callicebus";"torquatus";NA;39;12;0;4;NA;1245;3.85;NA;121.66;NA;1683.65;NA
+"Callimico_goeldii";"Callimico";"goeldii";NA;40;13;0;43;11.43;484;6.85;153.99;66.53;214.8;413.84;220.52
+"Callithrix_argentata";"Callithrix";"argentata";NA;41;14;0;16;7.95;345;9.5;NA;NA;201.6;701.52;NA
+"Callithrix_aurita";"Callithrix";"aurita";NA;42;14;0;NA;NA;429;6;140;NA;NA;NA;NA
+"Callithrix_emiliae";"Callithrix";"emiliae";NA;43;14;NA;NA;NA;309.58;NA;NA;NA;NA;NA;NA
+"Callithrix_geoffroyi";"Callithrix";"geoffroyi";NA;44;14;0;NA;NA;342;NA;NA;NA;NA;NA;NA
+"Callithrix_humeralifera";"Callithrix";"humeralifera";NA;45;14;0;4;NA;370;8.5;NA;99.01;180;NA;NA
+"Callithrix_jacchus";"Callithrix";"jacchus";NA;46;14;2;161;7.24;320;8.55;144;60.24;201.6;455.99;204.24
+"Callithrix_kuhli";"Callithrix";"kuhli";NA;47;14;0;NA;NA;374.99;NA;NA;NA;NA;NA;NA
+"Callithrix_mauesi";"Callithrix";"mauesi";NA;48;14;0;NA;NA;443.79;NA;NA;NA;NA;NA;NA
+"Callithrix_penicillata";"Callithrix";"penicillata";NA;49;14;0;NA;7.32;328;5.9;NA;NA;NA;NA;NA
+"Callithrix_pygmaea";"Callithrix";"pygmaea";NA;50;14;0;36;4.17;116;6;134.44;90.73;181.2;708.5;225.17
+"Cebus_albifrons";"Cebus";"albifrons";NA;51;15;1;13;65.45;2735;25;158.29;270.32;528;1501.69;428.61
+"Cebus_apella";"Cebus";"apella";NA;52;15;17;249;66.63;2936;7.9;154.99;263.12;541.2;1760.81;418.11
+"Cebus_capucinus";"Cebus";"capucinus";NA;53;15;5;60;72.93;2861;18.15;161.06;514.07;657.6;2134.73;675.13
+"Cebus_olivaceus";"Cebus";"olivaceus";NA;54;15;0;18;69.84;2931;11.45;NA;725.86;492;2525.48;NA
+"Cebus_xanthosternos";"Cebus";"xanthosternos";NA;55;15;NA;NA;66.09;2440;NA;NA;NA;NA;NA;NA
+"Cercocebus_agilis";"Cercocebus";"agilis";NA;56;16;NA;NA;116.43;7580;NA;NA;NA;NA;NA;NA
+"Cercocebus_galeritus";"Cercocebus";"galeritus";NA;57;16;0;19;99.07;7435;20.35;174.43;NA;252;2735.94;NA
+"Cercocebus_torquatus";"Cercocebus";"torquatus";NA;58;16;1;32;105.99;7485;26.85;168.98;NA;360;1318.86;NA
+"Cercocebus_torquatus_atys";"Cercocebus";"torquatus";"atys";59;16;NA;NA;94.68;8600;35;165.08;NA;321.6;1321.67;NA
+"Cercopithecus_albogularis";"Cercopithecus";"albogularis";NA;60;17;NA;NA;70.1;5620;32.5;NA;NA;NA;NA;NA
+"Cercopithecus_ascanius";"Cercopithecus";"ascanius";NA;61;17;1;26;59.58;3714;26.3;148.5;146.54;339.6;1718.73;295.04
+"Cercopithecus_campbelli";"Cercopithecus";"campbelli";NA;62;17;0;11;57.39;3600;11;180.8;362.93;396;NA;543.73
+"Cercopithecus_campbelli_lowei";"Cercopithecus";"campbelli";"lowei";63;17;NA;NA;55.64;3187;NA;NA;NA;NA;NA;NA
+"Cercopithecus_cephus";"Cercopithecus";"cephus";NA;64;17;0;8;65.26;3585;11;169.51;362.93;276;1521.9;532.44
+"Cercopithecus_cephus_cephus";"Cercopithecus";"cephus";"cephus";65;17;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Cercopithecus_cephus_ngottoensis";"Cercopithecus";"cephus";"ngottoensis";66;17;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Cercopithecus_diana";"Cercopithecus";"diana";NA;67;17;1;28;62.61;4550;24.95;NA;362.93;447.6;2279.95;NA
+"Cercopithecus_erythrogaster";"Cercopithecus";"erythrogaster";NA;68;17;0;3;NA;3444.88;NA;NA;NA;NA;NA;NA
+"Cercopithecus_erythrogaster_erythrogaster";"Cercopithecus";"erythrogaster";"erythrogaster";69;17;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Cercopithecus_erythrotis";"Cercopithecus";"erythrotis";NA;70;17;0;3;65.4;3250;NA;NA;NA;NA;NA;NA
+"Cercopithecus_hamlyni";"Cercopithecus";"hamlyni";NA;71;17;0;4;65.9;4425;NA;NA;NA;NA;NA;NA
+"Cercopithecus_lhoesti";"Cercopithecus";"lhoesti";NA;72;17;0;7;74.2;4710;17.4;NA;NA;192;NA;NA
+"Cercopithecus_mitis";"Cercopithecus";"mitis";NA;73;17;0;56;71.33;6109;16;138.39;688.08;325.2;2049.25;826.47
+"Cercopithecus_mona";"Cercopithecus";"mona";NA;74;17;0;8;61.84;3719;NA;NA;NA;360;NA;NA
+"Cercopithecus_neglectus";"Cercopithecus";"neglectus";NA;75;17;0;17;65.97;5450;4.5;172.07;417.62;315.6;2076.39;589.69
+"Cercopithecus_nictitans";"Cercopithecus";"nictitans";NA;76;17;0;7;71.13;5465;16;169.51;NA;276;1684.59;NA
+"Cercopithecus_petaurista";"Cercopithecus";"petaurista";NA;77;17;0;5;55.08;3609;14;NA;NA;228;NA;NA
+"Cercopithecus_pogonias";"Cercopithecus";"pogonias";NA;78;17;0;8;59.56;3580;15;169.51;NA;289.2;1684.59;NA
+"Cercopithecus_preussi";"Cercopithecus";"preussi";NA;79;17;0;2;NA;5132.57;3;NA;NA;NA;NA;NA
+"Cercopithecus_solatus";"Cercopithecus";"solatus";NA;80;17;0;6;NA;5256.91;10;NA;NA;NA;NA;NA
+"Cercopithecus_wolfi";"Cercopithecus";"wolfi";NA;81;17;0;7;61.45;3390;NA;NA;NA;NA;NA;NA
+"Cheirogaleus_crossleyi";"Cheirogaleus";"crossleyi";NA;82;18;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Cheirogaleus_major";"Cheirogaleus";"major";NA;83;18;0;3;5.81;400;1;70;47.14;180;420.91;117.14
+"Cheirogaleus_medius";"Cheirogaleus";"medius";NA;84;18;0;13;2.6;140;1;61.79;60.65;231.6;413.84;122.44
+"Chiropotes_satanas";"Chiropotes";"satanas";NA;85;19;0;21;48.33;3030;14.4;157.67;NA;216;NA;NA
+"Chlorocebus_aethiops";"Chlorocebus";"aethiops";NA;86;20;5;91;65;3720;NA;NA;217.76;379.2;NA;NA
+"Chlorocebus_pygerythrus";"Chlorocebus";"pygerythrus";NA;87;20;NA;NA;62.58;4324;NA;NA;NA;NA;NA;NA
+"Chlorocebus_pygerythrus_cynosurus";"Chlorocebus";"pygerythrus";"cynosurus";88;20;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Chlorocebus_sabaeus";"Chlorocebus";"sabaeus";NA;89;20;NA;NA;64.91;4312;NA;NA;NA;NA;NA;NA
+"Chlorocebus_tantalus";"Chlorocebus";"tantalus";NA;90;20;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Colobus_angolensis";"Colobus";"angolensis";NA;91;21;0;16;77.7;8625;10.9;NA;NA;NA;NA;NA
+"Colobus_angolensis_palliatus";"Colobus";"angolensis";"palliatus";92;21;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Colobus_guereza";"Colobus";"guereza";NA;93;21;0;42;74.39;8589;7.6;169.02;387.79;294;1929.19;556.81
+"Colobus_polykomos";"Colobus";"polykomos";NA;94;21;0;17;73.83;9100;10.2;172.69;213.78;366;1629.84;386.47
+"Colobus_satanas";"Colobus";"satanas";NA;95;21;0;10;74.9;8910;15.5;192.76;NA;NA;NA;NA
+"Colobus_vellerosus";"Colobus";"vellerosus";NA;96;21;NA;NA;73.07;7820;16;NA;NA;NA;NA;NA
+"Daubentonia_madagascariensis";"Daubentonia";"madagascariensis";NA;97;22;0;52;44.85;2555;1;166.48;197.7;291.6;834.72;364.18
+"Erythrocebus_patas";"Erythrocebus";"patas";NA;98;23;2;33;97.73;9450;28;167.2;211.79;286.8;1246.07;378.99
+"Eulemur_coronatus";"Eulemur";"coronatus";NA;99;24;0;11;20.65;1180;6.95;124.04;NA;220.8;701.52;NA
+"Eulemur_fulvus_albifrons";"Eulemur";"fulvus";"albifrons";100;24;NA;NA;21.45;2336;NA;NA;NA;NA;NA;NA
+"Eulemur_fulvus_albocollaris";"Eulemur";"fulvus";"albocollaris";101;24;NA;NA;22.1;2140;NA;NA;NA;NA;NA;NA
+"Eulemur_fulvus_collaris";"Eulemur";"fulvus";"collaris";102;24;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Eulemur_fulvus_fulvus";"Eulemur";"fulvus";"fulvus";103;24;1;81;25.77;2292;9.15;120.83;134.64;444;791.75;255.47
+"Eulemur_fulvus_mayottensis";"Eulemur";"fulvus";"mayottensis";104;24;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Eulemur_fulvus_rufus";"Eulemur";"fulvus";"rufus";105;24;NA;NA;25.4;2220;9.5;NA;NA;NA;NA;NA
+"Eulemur_fulvus_sanfordi";"Eulemur";"fulvus";"sanfordi";106;24;NA;NA;NA;2394.03;7.7;NA;NA;NA;NA;NA
+"Eulemur_macaco_flavifrons";"Eulemur";"macaco";"flavifrons";107;24;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Eulemur_macaco_macaco";"Eulemur";"macaco";"macaco";108;24;0;32;24.51;2390;9.2;127.49;143.28;360;660.75;270.77
+"Eulemur_mongoz";"Eulemur";"mongoz";NA;109;24;0;13;20.17;1212;2.7;129;151.13;360;1060.7;280.13
+"Eulemur_rubriventer";"Eulemur";"rubriventer";NA;110;24;0;13;26.23;1960;3.3;126.99;151.22;NA;566.36;278.21
+"Euoticus_elegantulus";"Euoticus";"elegantulus";NA;111;25;0;1;5.53;274;1;133.45;NA;180;NA;NA
+"Galago_alleni";"Galago";"alleni";NA;112;26;0;2;5.58;252;6;133;NA;144;283.18;NA
+"Galago_gallarum";"Galago";"gallarum";NA;113;26;NA;NA;NA;250;NA;NA;NA;NA;NA;NA
+"Galago_granti";"Galago";"granti";NA;114;26;0;NA;4.07;NA;NA;NA;NA;NA;NA;NA
+"Galago_matschiei";"Galago";"matschiei";NA;115;26;NA;NA;4.62;210;1;NA;NA;NA;NA;NA
+"Galago_moholi";"Galago";"moholi";NA;116;26;0;14;3.71;148;1;122.29;90.46;198;420.91;212.75
+"Galago_senegalensis";"Galago";"senegalensis";NA;117;26;0;20;3.96;194;3.5;126.98;93.93;204;330.37;220.91
+"Galagoides_demidoff";"Galagoides";"demidoff";NA;118;27;0;5;2.65;75;5.5;111;43.47;168;345.24;154.47
+"Galagoides_zanzibaricus";"Galagoides";"zanzibaricus";NA;119;27;0;NA;3.51;143;1;120;59.27;NA;322.75;179.27
+"Gorilla_beringei";"Gorilla";"beringei";NA;120;28;NA;NA;491.27;130000;NA;NA;NA;NA;NA;NA
+"Gorilla_gorilla_gorilla";"Gorilla";"gorilla";"gorilla";121;28;13;517;490.41;120950;6;257;920.35;648;3353.12;1177.35
+"Gorilla_gorilla_graueri";"Gorilla";"gorilla";"graueri";122;28;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Hapalemur_aureus";"Hapalemur";"aureus";NA;123;29;0;5;NA;1562.41;3;142.5;NA;NA;NA;NA
+"Hapalemur_griseus";"Hapalemur";"griseus";NA;124;29;0;40;14.09;709;3.1;141.24;136.29;205.2;1003.17;277.53
+"Hapalemur_griseus_alaotrensis";"Hapalemur";"griseus";"alaotrensis";125;29;NA;NA;13.8;1240;NA;NA;NA;NA;NA;NA
+"Hapalemur_griseus_griseus";"Hapalemur";"griseus";"griseus";126;29;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Hapalemur_griseus_meridionalis";"Hapalemur";"griseus";"meridionalis";127;29;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Hapalemur_griseus_occidentalis";"Hapalemur";"griseus";"occidentalis";128;29;NA;NA;14.2;NA;NA;NA;NA;NA;NA;NA
+"Hapalemur_simus";"Hapalemur";"simus";NA;129;29;0;8;27.14;2150;7.5;140;NA;144;NA;NA
+"Homo_sapiens";"Homo";"sapiens";NA;130;30;NA;NA;NA;58540.63;NA;274.78;725.86;1470;5582.93;1000.64
+"Homo_sapiens_neanderthalensis";"Homo";"sapiens";"neanderthalensis";131;30;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Hylobates_agilis";"Hylobates";"agilis";NA;132;31;0;16;91.16;5850;4.2;NA;NA;528;NA;NA
+"Hylobates_klossii";"Hylobates";"klossii";NA;133;31;0;4;87.99;5795;3;207.59;NA;NA;NA;NA
+"Hylobates_lar";"Hylobates";"lar";NA;134;31;0;86;101.87;5595;3.2;212.91;725.86;480;3852.57;938.77
+"Hylobates_moloch";"Hylobates";"moloch";NA;135;31;0;16;NA;5860.81;2.15;241.2;NA;NA;NA;NA
+"Hylobates_muelleri";"Hylobates";"muelleri";NA;136;31;0;5;85.13;5821;3.2;206.7;NA;NA;NA;NA
+"Hylobates_pileatus";"Hylobates";"pileatus";NA;137;31;0;16;84.69;5470;3.25;200.16;635.13;432;2454.24;835.29
+"Indri_indri";"Indri";"indri";NA;138;32;0;8;34.81;6335;3.1;136.5;331.34;NA;1605.69;467.84
+"Lagothrix_lagotricha";"Lagothrix";"lagotricha";NA;139;33;0;34;96.5;7150;33;223.99;312.66;360;1729.33;536.65
+"Lemur_catta";"Lemur";"catta";NA;140;34;4;103;22.9;2210;16.45;134.74;126.51;360;831.62;261.25
+"Leontopithecus_chrysomelas";"Leontopithecus";"chrysomelas";NA;141;35;0;46;11.84;655;6.7;NA;NA;NA;NA;NA
+"Leontopithecus_chrysopygus";"Leontopithecus";"chrysopygus";NA;142;35;0;38;NA;656.12;3.6;NA;NA;NA;NA;NA
+"Leontopithecus_rosalia";"Leontopithecus";"rosalia";NA;143;35;0;85;12.83;609;4.5;134;75.69;297.6;890.34;209.69
+"Lepilemur_aeeclis";"Lepilemur";"aeeclis";NA;144;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_ankaranensis";"Lepilemur";"ankaranensis";NA;145;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_dorsalis";"Lepilemur";"dorsalis";NA;146;36;0;1;6.7;870;1;NA;NA;NA;NA;NA
+"Lepilemur_edwardsi";"Lepilemur";"edwardsi";NA;147;36;0;5;7.25;931;1;NA;NA;NA;NA;NA
+"Lepilemur_hubbardorum";"Lepilemur";"hubbardorum";NA;148;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_leucopus";"Lepilemur";"leucopus";NA;149;36;0;2;6.87;606;1;135.92;121.66;103;620.76;257.58
+"Lepilemur_manasamody";"Lepilemur";"manasamody";NA;150;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_microdon";"Lepilemur";"microdon";NA;151;36;0;1;9.75;970;1;NA;NA;NA;NA;NA
+"Lepilemur_mitsinjoensis";"Lepilemur";"mitsinjoensis";NA;152;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_mustelinus";"Lepilemur";"mustelinus";NA;153;36;0;5;9.56;777;1;133.45;76.21;144;663.81;209.66
+"Lepilemur_otto";"Lepilemur";"otto";NA;154;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_randrianasoli";"Lepilemur";"randrianasoli";NA;155;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_ruficaudatus";"Lepilemur";"ruficaudatus";NA;156;36;0;2;8.25;805;1;135.92;119.32;NA;NA;255.24
+"Lepilemur_sahamalazensis";"Lepilemur";"sahamalazensis";NA;157;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_seali";"Lepilemur";"seali";NA;158;36;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Lepilemur_septentrionalis";"Lepilemur";"septentrionalis";NA;159;36;0;NA;NA;755.77;1;134.99;120.97;NA;377.57;255.96
+"Lophocebus_albigena";"Lophocebus";"albigena";NA;160;37;0;34;93.97;6950;16;182.64;211.71;392.4;2525.48;394.35
+"Lophocebus_aterrimus";"Lophocebus";"aterrimus";NA;161;37;0;6;101.59;6800;17.5;NA;NA;321.6;NA;NA
+"Loris_lydekkerianus";"Loris";"lydekkerianus";NA;162;38;NA;NA;6.34;267;NA;NA;NA;NA;NA;NA
+"Loris_tardigradus";"Loris";"tardigradus";NA;163;38;0;14;5.87;193;1;165.99;167.49;196.8;350.76;333.48
+"Macaca_arctoides";"Macaca";"arctoides";NA;164;39;1;48;100.7;10300;NA;176.6;377.66;360;1570.01;554.26
+"Macaca_assamensis";"Macaca";"assamensis";NA;165;39;0;17;90.46;9100;21;NA;NA;NA;NA;NA
+"Macaca_brunnescens";"Macaca";"brunnescens";NA;166;39;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Macaca_cyclopis";"Macaca";"cyclopis";NA;167;39;0;12;82;5470;20.2;161.06;205.24;NA;1650.01;366.3
+"Macaca_fascicularis";"Macaca";"fascicularis";NA;168;39;7;174;63.98;4251;27;164.69;283.53;456;1319.5;448.22
+"Macaca_fuscata";"Macaca";"fuscata";NA;169;39;45;253;102.92;9515;40.65;172.99;265.04;396;1460.77;438.03
+"Macaca_hecki";"Macaca";"hecki";NA;170;39;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Macaca_leonina";"Macaca";"leonina";NA;171;39;NA;NA;85.6;5642;NA;NA;NA;NA;NA;NA
+"Macaca_maura";"Macaca";"maura";NA;172;39;0;22;NA;7290.3;NA;167.19;497.16;NA;NA;664.35
+"Macaca_mulatta";"Macaca";"mulatta";NA;173;39;15;296;88.98;6793;38.5;166.07;304.16;432;1101.07;470.23
+"Macaca_munzala";"Macaca";"munzala";NA;174;39;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Macaca_nemestrina";"Macaca";"nemestrina";NA;175;39;3;51;105.59;8821;22.6;171;292.6;411.6;1427.17;463.6
+"Macaca_nemestrina_leonina";"Macaca";"nemestrina";"leonina";176;39;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Macaca_nemestrina_siberu";"Macaca";"nemestrina";"siberu";177;39;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Macaca_nigra";"Macaca";"nigra";NA;178;39;0;27;94.9;7680;35;172.43;365;216;1984.51;537.43
+"Macaca_nigrescens";"Macaca";"nigrescens";NA;179;39;NA;NA;NA;NA;14.5;NA;NA;NA;NA;NA
+"Macaca_ochreata";"Macaca";"ochreata";NA;180;39;0;3;NA;3400;NA;NA;NA;NA;NA;NA
+"Macaca_pagensis";"Macaca";"pagensis";NA;181;39;NA;NA;NA;4534.66;NA;NA;272.2;360;1227.12;NA
+"Macaca_radiata";"Macaca";"radiata";NA;182;39;0;34;74.87;5084;33.5;161.56;332.25;360;1785.78;493.81
+"Macaca_silenus";"Macaca";"silenus";NA;183;39;1;48;85;7500;21;172;362.93;480;1912.19;534.93
+"Macaca_sinica";"Macaca";"sinica";NA;184;39;0;12;69.7;4440;20.1;180.9;NA;420;1894.11;NA
+"Macaca_sylvanus";"Macaca";"sylvanus";NA;185;39;0;67;93.2;12078;18.3;164.84;210.25;264;1542.25;375.09
+"Macaca_thibetana";"Macaca";"thibetana";NA;186;39;1;42;NA;10593.06;21;169.02;451.79;NA;NA;620.81
+"Macaca_tonkeana";"Macaca";"tonkeana";NA;187;39;2;26;NA;10035.53;NA;NA;NA;NA;NA;NA
+"Mandrillus_leucophaeus";"Mandrillus";"leucophaeus";NA;188;40;0;18;148;15000;17;179.22;486.66;400.8;1745.96;665.88
+"Mandrillus_sphinx";"Mandrillus";"sphinx";NA;189;40;3;30;153.88;23600;13.9;173.99;348.01;555.96;2122.11;522
+"Microcebus_berthae";"Microcebus";"berthae";NA;190;41;NA;NA;NA;33.45;NA;NA;NA;NA;NA;NA
+"Microcebus_bongolavensis";"Microcebus";"bongolavensis";NA;191;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_danfossi";"Microcebus";"danfossi";NA;192;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_griseorufus";"Microcebus";"griseorufus";NA;193;41;NA;NA;NA;70.24;NA;NA;NA;NA;NA;NA
+"Microcebus_jollyae";"Microcebus";"jollyae";NA;194;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_lehilahytsara";"Microcebus";"lehilahytsara";NA;195;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_lokobensis";"Microcebus";"lokobensis";NA;196;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_macarthurii";"Microcebus";"macarthurii";NA;197;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_mamiratra";"Microcebus";"mamiratra";NA;198;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_mittermeieri";"Microcebus";"mittermeieri";NA;199;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_murinus";"Microcebus";"murinus";NA;200;41;0;66;1.63;65;1;60.34;40.45;186;355.53;100.79
+"Microcebus_myoxinus";"Microcebus";"myoxinus";NA;201;41;0;NA;NA;31.23;1;59.99;NA;NA;NA;NA
+"Microcebus_ravelobensis";"Microcebus";"ravelobensis";NA;202;41;0;NA;NA;58.6;NA;NA;NA;NA;NA;NA
+"Microcebus_rufus";"Microcebus";"rufus";NA;203;41;0;8;1.72;43;1;59.99;40;144;NA;99.99
+"Microcebus_sambiranensis";"Microcebus";"sambiranensis";NA;204;41;NA;NA;NA;49.06;NA;NA;NA;NA;NA;NA
+"Microcebus_simmonsi";"Microcebus";"simmonsi";NA;205;41;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Microcebus_tavaratra";"Microcebus";"tavaratra";NA;206;41;NA;NA;NA;68.01;NA;NA;NA;NA;NA;NA
+"Miopithecus_talapoin";"Miopithecus";"talapoin";NA;207;42;0;4;NA;1248.86;91.2;164.38;178.98;370.8;1733.36;343.36
+"Mirza_coquereli";"Mirza";"coquereli";NA;208;43;0;3;5.8;312;1;88.58;136;183.6;343.74;224.58
+"Mirza_zaza";"Mirza";"zaza";NA;209;43;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Nasalis_larvatus";"Nasalis";"larvatus";NA;210;44;0;17;92.3;14561;11.25;165.04;211.75;252;1894.11;376.79
+"Nomascus_concolor";"Nomascus";"concolor";NA;211;45;0;21;NA;6410.47;4;205.81;635.13;529.2;2454.24;840.94
+"Nomascus_gabriellae";"Nomascus";"gabriellae";NA;212;45;0;4;119.38;7365;1;NA;NA;NA;NA;NA
+"Nomascus_leucogenys";"Nomascus";"leucogenys";NA;213;45;0;8;NA;7320;1;NA;NA;NA;NA;NA
+"Nomascus_nasutus";"Nomascus";"nasutus";NA;214;45;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Nomascus_siki";"Nomascus";"siki";NA;215;45;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Nycticebus_bengalensis";"Nycticebus";"bengalensis";NA;216;46;NA;NA;13.49;1060;NA;NA;NA;NA;NA;NA
+"Nycticebus_coucang";"Nycticebus";"coucang";NA;217;46;0;37;10.13;653;1;191.09;181.21;318;660.82;372.3
+"Nycticebus_javanicus";"Nycticebus";"javanicus";NA;218;46;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Nycticebus_menagensis";"Nycticebus";"menagensis";NA;219;46;NA;NA;9.67;634;NA;NA;NA;NA;NA;NA
+"Nycticebus_pygmaeus";"Nycticebus";"pygmaeus";NA;220;46;0;19;7.23;307;1;185.42;NA;NA;NA;NA
+"Otolemur_crassicaudatus";"Otolemur";"crassicaudatus";NA;221;47;1;36;11.78;1150;3.5;131.04;124.62;225.6;609.86;255.66
+"Otolemur_garnettii";"Otolemur";"garnettii";NA;222;47;1;12;11.5;764;1;132.24;139.2;204;592.15;271.44
+"Pan_paniscus";"Pan";"paniscus";NA;223;48;5;225;341.29;39100;85;235.24;1081.31;576;5465.72;1316.55
+"Pan_troglodytes_schweinfurthii";"Pan";"troglodytes";"schweinfurthii";224;48;NA;NA;390.33;38200;NA;NA;NA;NA;NA;NA
+"Pan_troglodytes_troglodytes";"Pan";"troglodytes";"troglodytes";225;48;214;755;363.05;52750;50;231.49;1260.81;720;3897.96;1492.3
+"Pan_troglodytes_vellerosus";"Pan";"troglodytes";"vellerosus";226;48;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Pan_troglodytes_verus";"Pan";"troglodytes";"verus";227;48;NA;NA;371.74;43950;NA;NA;NA;NA;NA;NA
+"Papio_anubis";"Papio";"anubis";NA;228;49;4;43;167.42;18150;40;178.96;596.6;302.4;NA;775.56
+"Papio_cynocephalus";"Papio";"cynocephalus";NA;229;49;2;114;163.19;17150;48.2;172.99;450.42;540;2560.56;623.41
+"Papio_hamadryas";"Papio";"hamadryas";NA;230;49;1;78;146.17;14150;36.9;180;363.96;450;1652.37;543.96
+"Papio_papio";"Papio";"papio";NA;231;49;3;8;142.5;18026.05;NA;184.42;NA;480;NA;NA
+"Papio_ursinus";"Papio";"ursinus";NA;232;49;5;22;178;22300;47;185.92;877.09;540;1543.35;1063.01
+"Perodicticus_potto";"Perodicticus";"potto";NA;233;50;0;10;12.42;835;1;193;149.15;312;561.58;342.15
+"Phaner_furcifer";"Phaner";"furcifer";NA;234;51;0;1;NA;409.87;1;174.46;NA;144;NA;NA
+"Phaner_furcifer_pallescens";"Phaner";"furcifer";"pallescens";235;51;NA;NA;6.68;339;NA;NA;NA;NA;NA;NA
+"Piliocolobus_badius";"Piliocolobus";"badius";NA;236;52;0;52;63.59;8285;34;151.41;783.93;NA;1473.2;935.34
+"Piliocolobus_foai";"Piliocolobus";"foai";NA;237;52;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Piliocolobus_gordonorum";"Piliocolobus";"gordonorum";NA;238;52;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Piliocolobus_kirkii";"Piliocolobus";"kirkii";NA;239;52;1;7;57.25;5630;33.6;165;NA;NA;NA;NA
+"Piliocolobus_pennantii";"Piliocolobus";"pennantii";NA;240;52;0;NA;NA;10896;NA;NA;NA;NA;NA;NA
+"Piliocolobus_preussi";"Piliocolobus";"preussi";NA;241;52;0;NA;NA;8865.71;40;195;NA;NA;NA;NA
+"Piliocolobus_rufomitratus";"Piliocolobus";"rufomitratus";NA;242;52;NA;NA;NA;8030.75;24.5;195;NA;NA;NA;NA
+"Piliocolobus_tephrosceles";"Piliocolobus";"tephrosceles";NA;243;52;NA;NA;70.95;8409;34;NA;NA;NA;NA;NA
+"Piliocolobus_tholloni";"Piliocolobus";"tholloni";NA;244;52;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Pithecia_irrorata";"Pithecia";"irrorata";NA;245;53;0;7;NA;2308.17;4.4;NA;NA;NA;NA;NA
+"Pithecia_pithecia";"Pithecia";"pithecia";NA;246;53;0;28;32.26;1760;2.7;161.13;113.15;248.4;1089.37;274.28
+"Pongo_abelii";"Pongo";"abelii";NA;247;54;NA;NA;389.5;62815;NA;NA;NA;NA;NA;NA
+"Pongo_pygmaeus";"Pongo";"pygmaeus";NA;248;54;86;321;377.38;58542;1;259.42;1088.8;720;3318.62;1348.22
+"Presbytis_comata";"Presbytis";"comata";NA;249;55;0;11;80.3;6695;7.05;NA;NA;NA;NA;NA
+"Presbytis_melalophos";"Presbytis";"melalophos";NA;250;55;0;6;64.85;6560;14;NA;NA;192;NA;NA
+"Procolobus_verus";"Procolobus";"verus";NA;251;56;0;3;52.6;4450;6.3;167.84;NA;NA;NA;NA
+"Propithecus_coquereli";"Propithecus";"coquereli";NA;252;57;NA;NA;30.19;3729;5.5;140.99;180.96;NA;NA;321.95
+"Propithecus_deckenii";"Propithecus";"deckenii";NA;253;57;NA;NA;30.15;3532;NA;NA;NA;NA;NA;NA
+"Propithecus_diadema";"Propithecus";"diadema";NA;254;57;0;28;39.8;6130;4.95;152.08;256.27;NA;1683.65;408.35
+"Propithecus_edwardsi";"Propithecus";"edwardsi";NA;255;57;NA;NA;39.49;5682;6;NA;NA;NA;NA;NA
+"Propithecus_tattersalli";"Propithecus";"tattersalli";NA;256;57;0;9;NA;3531.39;4.1;NA;152.13;NA;NA;NA
+"Propithecus_verreauxi";"Propithecus";"verreauxi";NA;257;57;1;41;26.21;2955;6.3;149.77;177.83;247.2;943.94;327.6
+"Pygathrix_cinerea";"Pygathrix";"cinerea";NA;258;58;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Pygathrix_nemaeus";"Pygathrix";"nemaeus";NA;259;58;0;25;91.41;9720;9.3;182.88;NA;300;NA;NA
+"Rhinopithecus_avunculus";"Rhinopithecus";"avunculus";NA;260;59;0;11;NA;9086.19;30;200;NA;NA;NA;NA
+"Rhinopithecus_bieti";"Rhinopithecus";"bieti";NA;261;59;0;NA;NA;11000.54;50;170;NA;NA;755.15;NA
+"Rhinopithecus_brelichi";"Rhinopithecus";"brelichi";NA;262;59;0;16;NA;12267.15;NA;200;NA;NA;NA;NA
+"Rhinopithecus_roxellana";"Rhinopithecus";"roxellana";NA;263;59;0;36;117.76;14750;65;199.34;NA;NA;NA;NA
+"Rungwecebus_kipunji";"Rungwecebus";"kipunji";NA;264;60;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Saguinus_bicolor";"Saguinus";"bicolor";NA;265;61;0;9;NA;465;6.7;158.16;NA;NA;NA;NA
+"Saguinus_fuscicollis";"Saguinus";"fuscicollis";NA;266;61;2;81;7.94;401;6;148;90.1;294;406.61;238.1
+"Saguinus_fuscicollis_melanoleucus";"Saguinus";"fuscicollis";"melanoleucus";267;61;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Saguinus_geoffroyi";"Saguinus";"geoffroyi";NA;268;61;0;NA;10.14;517;6.9;NA;NA;NA;NA;NA
+"Saguinus_imperator";"Saguinus";"imperator";NA;269;61;0;16;NA;407.91;5;NA;NA;242.4;NA;NA
+"Saguinus_leucopus";"Saguinus";"leucopus";NA;270;61;0;3;9.7;525;7.5;142.5;NA;NA;NA;NA
+"Saguinus_midas";"Saguinus";"midas";NA;271;61;0;17;9.78;563;5.55;138.24;69.6;184.8;841.82;207.84
+"Saguinus_mystax";"Saguinus";"mystax";NA;272;61;0;46;11.09;584;5.4;148.28;NA;NA;556.85;NA
+"Saguinus_niger";"Saguinus";"niger";NA;273;61;NA;NA;9.48;375;NA;NA;NA;NA;NA;NA
+"Saguinus_oedipus";"Saguinus";"oedipus";NA;274;61;0;153;9.76;431;7.05;166.49;49.85;277.2;680.38;216.34
+"Saguinus_tripartitus";"Saguinus";"tripartitus";NA;275;61;0;5;NA;385.05;NA;NA;NA;NA;NA;NA
+"Saimiri_boliviensis";"Saimiri";"boliviensis";NA;276;62;0;36;NA;799.45;60;157.79;NA;NA;NA;NA
+"Saimiri_oerstedii";"Saimiri";"oerstedii";NA;277;62;1;4;25.07;789;25.1;161;362.93;NA;NA;523.93
+"Saimiri_sciureus";"Saimiri";"sciureus";NA;278;62;1;89;24.14;799;34.85;164.09;177.41;324;1399.88;341.5
+"Saimiri_ustus";"Saimiri";"ustus";NA;279;62;0;4;NA;886.47;NA;NA;238.64;NA;NA;NA
+"Semnopithecus_entellus";"Semnopithecus";"entellus";NA;280;63;2;98;110.93;14742;19;197.7;402.1;300;1497.64;599.8
+"Symphalangus_syndactylus";"Symphalangus";"syndactylus";NA;281;64;0;40;123.5;11295;3.8;230.66;635.38;456;3788.23;866.04
+"Tarsius_bancanus";"Tarsius";"bancanus";NA;282;65;0;8;3.16;126;1;125.84;78.55;144;658.68;204.39
+"Tarsius_dentatus";"Tarsius";"dentatus";NA;283;65;0;2;3;113;1;NA;NA;NA;NA;NA
+"Tarsius_lariang";"Tarsius";"lariang";NA;284;65;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Tarsius_syrichta";"Tarsius";"syrichta";NA;285;65;0;10;3.36;126;1;177.99;82.49;180;NA;260.48
+"Theropithecus_gelada";"Theropithecus";"gelada";NA;286;66;0;34;133.33;15350;10;178.64;494.95;336;1894.11;673.59
+"Trachypithecus_auratus";"Trachypithecus";"auratus";NA;287;67;0;2;NA;9719.6;11;NA;NA;NA;NA;NA
+"Trachypithecus_cristatus";"Trachypithecus";"cristatus";NA;288;67;0;8;57.86;6394;27.4;NA;362.93;373.2;NA;NA
+"Trachypithecus_delacouri";"Trachypithecus";"delacouri";NA;289;67;0;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Trachypithecus_francoisi";"Trachypithecus";"francoisi";NA;290;67;0;45;NA;8139.93;NA;NA;391.76;NA;NA;NA
+"Trachypithecus_geei";"Trachypithecus";"geei";NA;291;67;0;7;81.3;10150;11;NA;NA;NA;NA;NA
+"Trachypithecus_germaini";"Trachypithecus";"germaini";NA;292;67;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Trachypithecus_johnii";"Trachypithecus";"johnii";NA;293;67;1;9;84.6;11600;10;NA;NA;NA;NA;NA
+"Trachypithecus_laotum";"Trachypithecus";"laotum";NA;294;67;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Trachypithecus_obscurus";"Trachypithecus";"obscurus";NA;295;67;0;6;62.12;7056;10;146.63;362.93;300;NA;509.56
+"Trachypithecus_phayrei";"Trachypithecus";"phayrei";NA;296;67;0;16;72.84;7475;12.9;180.61;305.87;NA;NA;486.48
+"Trachypithecus_pileatus";"Trachypithecus";"pileatus";NA;297;67;0;5;103.64;11794;8.5;NA;NA;NA;NA;NA
+"Trachypithecus_poliocephalus";"Trachypithecus";"poliocephalus";NA;298;67;NA;NA;NA;NA;NA;NA;NA;NA;NA;NA
+"Trachypithecus_vetulus";"Trachypithecus";"vetulus";NA;299;67;0;2;61.29;6237;8.35;204.72;245.78;276;1113.7;450.5
+"Varecia_rubra";"Varecia";"rubra";NA;300;68;NA;NA;31.08;3470;NA;NA;NA;NA;NA;NA
+"Varecia_variegata_variegata";"Varecia";"variegata";"variegata";301;68;0;57;32.12;3575;2.8;102.5;90.73;384;701.52;193.23
diff --git a/Rethinking_2/Data/cars.csv b/Rethinking_2/Data/cars.csv
new file mode 100644
index 0000000..34b3d46
--- /dev/null
+++ b/Rethinking_2/Data/cars.csv
@@ -0,0 +1,51 @@
+"","speed","dist"
+"1",4,2
+"2",4,10
+"3",7,4
+"4",7,22
+"5",8,16
+"6",9,10
+"7",10,18
+"8",10,26
+"9",10,34
+"10",11,17
+"11",11,28
+"12",12,14
+"13",12,20
+"14",12,24
+"15",12,28
+"16",13,26
+"17",13,34
+"18",13,34
+"19",13,46
+"20",14,26
+"21",14,36
+"22",14,60
+"23",14,80
+"24",15,20
+"25",15,26
+"26",15,54
+"27",16,32
+"28",16,40
+"29",17,32
+"30",17,40
+"31",17,50
+"32",18,42
+"33",18,56
+"34",18,76
+"35",18,84
+"36",19,36
+"37",19,46
+"38",19,68
+"39",20,32
+"40",20,48
+"41",20,52
+"42",20,56
+"43",20,64
+"44",22,66
+"45",23,54
+"46",24,70
+"47",24,92
+"48",24,93
+"49",24,120
+"50",25,85