{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "532c6736",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n"
     ]
    }
   ],
   "source": [
    "import arviz as az\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pymc as pm\n",
    "\n",
    "import causalpy as cp\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1b3aa75",
   "metadata": {},
   "source": [
    "## Variable Selection Priors and Instrumental Variable Designs\n",
    "\n",
    "When building causal inference models, we often face a dilemma: we want to control for confounders to get unbiased causal estimates, but we're not always certain which variables are the true confounders. Include too few, and we risk omitted variable bias. Include too many, and we introduce noise that inflates our uncertainty or, worse, creates multicollinearity that destabilizes our estimates.\n",
    "\n",
    "Traditional approaches force us to make hard choices upfront—which variables to include, which to exclude. This in ideal cases should be driven by theory. But what if we could let the data help us make these decisions while still maintaining the principled probabilistic framework of Bayesian inference? This is where variable selection priors come in. Let's first simulate some data with some natural confounding structure. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "046aa8e0",
   "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>Y_cont</th>\n",
       "      <th>Y_bin</th>\n",
       "      <th>T_cont</th>\n",
       "      <th>T_bin</th>\n",
       "      <th>alpha</th>\n",
       "      <th>feature_0</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>feature_4</th>\n",
       "      <th>...</th>\n",
       "      <th>feature_8</th>\n",
       "      <th>feature_9</th>\n",
       "      <th>feature_10</th>\n",
       "      <th>feature_11</th>\n",
       "      <th>feature_12</th>\n",
       "      <th>feature_13</th>\n",
       "      <th>Y_cont_scaled</th>\n",
       "      <th>Y_bin_scaled</th>\n",
       "      <th>T_cont_scaled</th>\n",
       "      <th>T_bin_scaled</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.169837</td>\n",
       "      <td>-0.170346</td>\n",
       "      <td>1.113394</td>\n",
       "      <td>0</td>\n",
       "      <td>9.236102</td>\n",
       "      <td>1.294441</td>\n",
       "      <td>0.418241</td>\n",
       "      <td>0.536286</td>\n",
       "      <td>-0.615573</td>\n",
       "      <td>-1.173784</td>\n",
       "      <td>...</td>\n",
       "      <td>0.559393</td>\n",
       "      <td>1.111766</td>\n",
       "      <td>-0.216069</td>\n",
       "      <td>0.451496</td>\n",
       "      <td>-0.863189</td>\n",
       "      <td>0.319180</td>\n",
       "      <td>0.268475</td>\n",
       "      <td>-0.438362</td>\n",
       "      <td>0.347809</td>\n",
       "      <td>-1.022452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.479049</td>\n",
       "      <td>6.662990</td>\n",
       "      <td>2.272020</td>\n",
       "      <td>1</td>\n",
       "      <td>3.787487</td>\n",
       "      <td>-0.885005</td>\n",
       "      <td>0.315013</td>\n",
       "      <td>0.810138</td>\n",
       "      <td>1.137214</td>\n",
       "      <td>0.203685</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.017179</td>\n",
       "      <td>0.410818</td>\n",
       "      <td>0.670074</td>\n",
       "      <td>1.954944</td>\n",
       "      <td>0.888255</td>\n",
       "      <td>-0.506518</td>\n",
       "      <td>0.916637</td>\n",
       "      <td>1.321011</td>\n",
       "      <td>0.726014</td>\n",
       "      <td>0.977650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.307821</td>\n",
       "      <td>4.118982</td>\n",
       "      <td>2.062946</td>\n",
       "      <td>1</td>\n",
       "      <td>3.311157</td>\n",
       "      <td>-1.075537</td>\n",
       "      <td>1.058536</td>\n",
       "      <td>1.908944</td>\n",
       "      <td>1.122914</td>\n",
       "      <td>0.611691</td>\n",
       "      <td>...</td>\n",
       "      <td>0.050641</td>\n",
       "      <td>1.245489</td>\n",
       "      <td>-1.070642</td>\n",
       "      <td>-0.060250</td>\n",
       "      <td>-1.857638</td>\n",
       "      <td>0.806913</td>\n",
       "      <td>0.635421</td>\n",
       "      <td>0.666008</td>\n",
       "      <td>0.657767</td>\n",
       "      <td>0.977650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9.781360</td>\n",
       "      <td>0.649647</td>\n",
       "      <td>4.043904</td>\n",
       "      <td>1</td>\n",
       "      <td>8.423450</td>\n",
       "      <td>0.969380</td>\n",
       "      <td>0.698199</td>\n",
       "      <td>0.314419</td>\n",
       "      <td>1.446987</td>\n",
       "      <td>-2.729092</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.052429</td>\n",
       "      <td>1.143079</td>\n",
       "      <td>-1.757701</td>\n",
       "      <td>-1.167276</td>\n",
       "      <td>-0.164620</td>\n",
       "      <td>1.687004</td>\n",
       "      <td>0.854768</td>\n",
       "      <td>-0.227239</td>\n",
       "      <td>1.304402</td>\n",
       "      <td>0.977650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.739283</td>\n",
       "      <td>6.812030</td>\n",
       "      <td>0.642417</td>\n",
       "      <td>1</td>\n",
       "      <td>6.613922</td>\n",
       "      <td>0.245569</td>\n",
       "      <td>-0.338491</td>\n",
       "      <td>0.814305</td>\n",
       "      <td>-0.859798</td>\n",
       "      <td>0.334968</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.547622</td>\n",
       "      <td>0.778724</td>\n",
       "      <td>0.678956</td>\n",
       "      <td>1.715229</td>\n",
       "      <td>-0.439130</td>\n",
       "      <td>-0.238847</td>\n",
       "      <td>0.496327</td>\n",
       "      <td>1.359385</td>\n",
       "      <td>0.194070</td>\n",
       "      <td>0.977650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2495</th>\n",
       "      <td>-5.099912</td>\n",
       "      <td>-0.870746</td>\n",
       "      <td>-1.409722</td>\n",
       "      <td>0</td>\n",
       "      <td>6.565630</td>\n",
       "      <td>0.226252</td>\n",
       "      <td>1.589653</td>\n",
       "      <td>0.056005</td>\n",
       "      <td>-0.386026</td>\n",
       "      <td>-0.462251</td>\n",
       "      <td>...</td>\n",
       "      <td>0.987633</td>\n",
       "      <td>0.246870</td>\n",
       "      <td>-0.202917</td>\n",
       "      <td>0.178579</td>\n",
       "      <td>0.763186</td>\n",
       "      <td>0.527462</td>\n",
       "      <td>-0.464865</td>\n",
       "      <td>-0.618693</td>\n",
       "      <td>-0.475800</td>\n",
       "      <td>-1.022452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2496</th>\n",
       "      <td>-32.742858</td>\n",
       "      <td>-7.337551</td>\n",
       "      <td>-8.468435</td>\n",
       "      <td>0</td>\n",
       "      <td>4.760520</td>\n",
       "      <td>-0.495792</td>\n",
       "      <td>0.546002</td>\n",
       "      <td>0.209072</td>\n",
       "      <td>0.666614</td>\n",
       "      <td>0.400847</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.798775</td>\n",
       "      <td>-0.616483</td>\n",
       "      <td>0.431552</td>\n",
       "      <td>1.238957</td>\n",
       "      <td>0.957759</td>\n",
       "      <td>-0.583051</td>\n",
       "      <td>-2.916171</td>\n",
       "      <td>-2.283696</td>\n",
       "      <td>-2.779944</td>\n",
       "      <td>-1.022452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2497</th>\n",
       "      <td>6.759804</td>\n",
       "      <td>1.912040</td>\n",
       "      <td>1.615921</td>\n",
       "      <td>0</td>\n",
       "      <td>10.445934</td>\n",
       "      <td>1.778373</td>\n",
       "      <td>0.097808</td>\n",
       "      <td>-0.807658</td>\n",
       "      <td>0.380358</td>\n",
       "      <td>-0.455391</td>\n",
       "      <td>...</td>\n",
       "      <td>0.668040</td>\n",
       "      <td>1.471963</td>\n",
       "      <td>0.573966</td>\n",
       "      <td>-0.288768</td>\n",
       "      <td>-0.861025</td>\n",
       "      <td>0.372657</td>\n",
       "      <td>0.586824</td>\n",
       "      <td>0.097788</td>\n",
       "      <td>0.511847</td>\n",
       "      <td>-1.022452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2498</th>\n",
       "      <td>-11.249395</td>\n",
       "      <td>-1.938808</td>\n",
       "      <td>-3.103529</td>\n",
       "      <td>0</td>\n",
       "      <td>5.715321</td>\n",
       "      <td>-0.113872</td>\n",
       "      <td>0.747480</td>\n",
       "      <td>1.635159</td>\n",
       "      <td>-1.136585</td>\n",
       "      <td>-0.007239</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.404202</td>\n",
       "      <td>-2.074570</td>\n",
       "      <td>0.022878</td>\n",
       "      <td>1.345018</td>\n",
       "      <td>0.705361</td>\n",
       "      <td>0.414329</td>\n",
       "      <td>-1.010186</td>\n",
       "      <td>-0.893686</td>\n",
       "      <td>-1.028702</td>\n",
       "      <td>-1.022452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2499</th>\n",
       "      <td>21.658258</td>\n",
       "      <td>7.675401</td>\n",
       "      <td>5.660952</td>\n",
       "      <td>1</td>\n",
       "      <td>5.741192</td>\n",
       "      <td>-0.103523</td>\n",
       "      <td>-1.083828</td>\n",
       "      <td>0.896827</td>\n",
       "      <td>0.146243</td>\n",
       "      <td>1.363973</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.310958</td>\n",
       "      <td>-0.004224</td>\n",
       "      <td>0.206620</td>\n",
       "      <td>0.012787</td>\n",
       "      <td>-1.777783</td>\n",
       "      <td>0.340176</td>\n",
       "      <td>1.907981</td>\n",
       "      <td>1.581676</td>\n",
       "      <td>1.832247</td>\n",
       "      <td>0.977650</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2500 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Y_cont     Y_bin    T_cont  T_bin      alpha  feature_0  feature_1  \\\n",
       "0      3.169837 -0.170346  1.113394      0   9.236102   1.294441   0.418241   \n",
       "1     10.479049  6.662990  2.272020      1   3.787487  -0.885005   0.315013   \n",
       "2      7.307821  4.118982  2.062946      1   3.311157  -1.075537   1.058536   \n",
       "3      9.781360  0.649647  4.043904      1   8.423450   0.969380   0.698199   \n",
       "4      5.739283  6.812030  0.642417      1   6.613922   0.245569  -0.338491   \n",
       "...         ...       ...       ...    ...        ...        ...        ...   \n",
       "2495  -5.099912 -0.870746 -1.409722      0   6.565630   0.226252   1.589653   \n",
       "2496 -32.742858 -7.337551 -8.468435      0   4.760520  -0.495792   0.546002   \n",
       "2497   6.759804  1.912040  1.615921      0  10.445934   1.778373   0.097808   \n",
       "2498 -11.249395 -1.938808 -3.103529      0   5.715321  -0.113872   0.747480   \n",
       "2499  21.658258  7.675401  5.660952      1   5.741192  -0.103523  -1.083828   \n",
       "\n",
       "      feature_2  feature_3  feature_4  ...  feature_8  feature_9  feature_10  \\\n",
       "0      0.536286  -0.615573  -1.173784  ...   0.559393   1.111766   -0.216069   \n",
       "1      0.810138   1.137214   0.203685  ...  -0.017179   0.410818    0.670074   \n",
       "2      1.908944   1.122914   0.611691  ...   0.050641   1.245489   -1.070642   \n",
       "3      0.314419   1.446987  -2.729092  ...  -1.052429   1.143079   -1.757701   \n",
       "4      0.814305  -0.859798   0.334968  ...  -0.547622   0.778724    0.678956   \n",
       "...         ...        ...        ...  ...        ...        ...         ...   \n",
       "2495   0.056005  -0.386026  -0.462251  ...   0.987633   0.246870   -0.202917   \n",
       "2496   0.209072   0.666614   0.400847  ...  -0.798775  -0.616483    0.431552   \n",
       "2497  -0.807658   0.380358  -0.455391  ...   0.668040   1.471963    0.573966   \n",
       "2498   1.635159  -1.136585  -0.007239  ...  -2.404202  -2.074570    0.022878   \n",
       "2499   0.896827   0.146243   1.363973  ...  -1.310958  -0.004224    0.206620   \n",
       "\n",
       "      feature_11  feature_12  feature_13  Y_cont_scaled  Y_bin_scaled  \\\n",
       "0       0.451496   -0.863189    0.319180       0.268475     -0.438362   \n",
       "1       1.954944    0.888255   -0.506518       0.916637      1.321011   \n",
       "2      -0.060250   -1.857638    0.806913       0.635421      0.666008   \n",
       "3      -1.167276   -0.164620    1.687004       0.854768     -0.227239   \n",
       "4       1.715229   -0.439130   -0.238847       0.496327      1.359385   \n",
       "...          ...         ...         ...            ...           ...   \n",
       "2495    0.178579    0.763186    0.527462      -0.464865     -0.618693   \n",
       "2496    1.238957    0.957759   -0.583051      -2.916171     -2.283696   \n",
       "2497   -0.288768   -0.861025    0.372657       0.586824      0.097788   \n",
       "2498    1.345018    0.705361    0.414329      -1.010186     -0.893686   \n",
       "2499    0.012787   -1.777783    0.340176       1.907981      1.581676   \n",
       "\n",
       "      T_cont_scaled  T_bin_scaled  \n",
       "0          0.347809     -1.022452  \n",
       "1          0.726014      0.977650  \n",
       "2          0.657767      0.977650  \n",
       "3          1.304402      0.977650  \n",
       "4          0.194070      0.977650  \n",
       "...             ...           ...  \n",
       "2495      -0.475800     -1.022452  \n",
       "2496      -2.779944     -1.022452  \n",
       "2497       0.511847     -1.022452  \n",
       "2498      -1.028702     -1.022452  \n",
       "2499       1.832247      0.977650  \n",
       "\n",
       "[2500 rows x 23 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def inv_logit(z):\n",
    "    \"\"\"Compute the inverse logit (sigmoid) of z.\"\"\"\n",
    "    return 1 / (1 + np.exp(-z))\n",
    "\n",
    "\n",
    "def simulate_data(n=2500, alpha_true=3.0, rho=0.6, cate_estimation=False):\n",
    "    # Exclusion restrictions:\n",
    "    # X[0], X[1] affect both Y and T (confounders)\n",
    "    # X[2], X[3] affect ONLY T (instruments for T)\n",
    "    # X[4] affects ONLY Y (predictor of Y only)\n",
    "\n",
    "    betaY = np.array(\n",
    "        [0.5, -0.3, 0.0, 0.0, 0.4, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
    "    )  # X[2], X[3] excluded\n",
    "    betaD = np.array(\n",
    "        [0.7, 0.1, -0.4, 0.3, 0.0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
    "    )  # X[4] excluded\n",
    "    p = len(betaY)\n",
    "\n",
    "    # noise variances and correlation\n",
    "    sigma_U = 3.0\n",
    "    sigma_V = 3.0\n",
    "\n",
    "    # design matrix (n × p) with mean-zero columns\n",
    "    X = np.random.normal(size=(n, p))\n",
    "    X = (X - X.mean(axis=0)) / X.std(axis=0)\n",
    "\n",
    "    mean = [0, 0]\n",
    "    cov = [[sigma_U**2, rho * sigma_U * sigma_V], [rho * sigma_U * sigma_V, sigma_V**2]]\n",
    "    errors = np.random.multivariate_normal(mean, cov, size=n)\n",
    "    U = errors[:, 0]  # error in outcome equation\n",
    "    V = errors[:, 1]  #\n",
    "\n",
    "    # continuous treatment\n",
    "    T_cont = X @ betaD + V\n",
    "\n",
    "    # latent variable for binary treatment\n",
    "    T_latent = X @ betaD + V\n",
    "    T_bin = np.random.binomial(n=1, p=inv_logit(T_latent), size=n)\n",
    "\n",
    "    alpha_individual = 3.0 + 2.5 * X[:, 0]\n",
    "\n",
    "    # outcomes\n",
    "    Y_cont = alpha_true * T_cont + X @ betaY + U\n",
    "    if cate_estimation:\n",
    "        Y_bin = alpha_individual * T_bin + X @ betaY + U\n",
    "    else:\n",
    "        Y_bin = alpha_true * T_bin + X @ betaY + U\n",
    "\n",
    "    # combine into DataFrame\n",
    "    data = pd.DataFrame(\n",
    "        {\n",
    "            \"Y_cont\": Y_cont,\n",
    "            \"Y_bin\": Y_bin,\n",
    "            \"T_cont\": T_cont,\n",
    "            \"T_bin\": T_bin,\n",
    "        }\n",
    "    )\n",
    "    data[\"alpha\"] = alpha_true + alpha_individual\n",
    "    for j in range(p):\n",
    "        data[f\"feature_{j}\"] = X[:, j]\n",
    "    data[\"Y_cont_scaled\"] = (data[\"Y_cont\"] - data[\"Y_cont\"].mean()) / data[\n",
    "        \"Y_cont\"\n",
    "    ].std(ddof=1)\n",
    "    data[\"Y_bin_scaled\"] = (data[\"Y_bin\"] - data[\"Y_bin\"].mean()) / data[\"Y_bin\"].std(\n",
    "        ddof=1\n",
    "    )\n",
    "    data[\"T_cont_scaled\"] = (data[\"T_cont\"] - data[\"T_cont\"].mean()) / data[\n",
    "        \"T_cont\"\n",
    "    ].std(ddof=1)\n",
    "    data[\"T_bin_scaled\"] = (data[\"T_bin\"] - data[\"T_bin\"].mean()) / data[\"T_bin\"].std(\n",
    "        ddof=1\n",
    "    )\n",
    "    return data\n",
    "\n",
    "\n",
    "data = simulate_data()\n",
    "instruments_data = data.copy()\n",
    "features = [col for col in data.columns if \"feature\" in col]\n",
    "formula = \"Y_cont ~ T_cont + \" + \" + \".join(features)\n",
    "instruments_formula = \"T_cont ~ 1 + \" + \" + \".join(features)\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e2472e18",
   "metadata": {},
   "source": [
    "CausalPy's  `Variable Selection` module provides a way to encode our uncertainty about variable relevance directly into the prior distribution. Rather than choosing which predictors to include, we specify priors that allow coefficients to be shrunk toward zero (or exactly zero) when the data doesn't support their inclusion. The key insight is that variable selection becomes part of the inference problem rather than a preprocessing step. The module offers two fundamentally different approaches to variable selection, each reflecting a different belief about how sparsity manifests in the world.\n",
    "\n",
    "### The Spike-and-Slab: Discrete Choices\n",
    "\n",
    "The spike-and-slab prior embodies a binary worldview: each variable either matters or it doesn't. Mathematically, we express this as:\n",
    "\n",
    "$$ \\beta_{j} = \\gamma_{j} \\cdot \\beta_{j_\\text{raw}}$$\n",
    "\n",
    "such that  \n",
    "\n",
    "$$ \\gamma_{j} \\in \\{0, 1\\}$$\n",
    "\n",
    "So we have the \"spike\"—the coefficient is exactly zero. When $\\gamma_{j} = 1$, we have the \"slab\" i.e. the coefficient takes on a value from the raw distribution.\n",
    "This approach appeals to our intuition about many real-world scenarios. Consider a propensity score model predicting whether someone receives a treatment. Some demographic variables might genuinely have no relationship with treatment assignment, while others are strongly predictive. The spike-and-slab says: let's let each variable clearly declare itself as relevant or irrelevant.\n",
    "\n",
    "### The Regularised Horseshoe: Gentle Moderation\n",
    "\n",
    "The horseshoe prior takes a different philosophical stance. Instead of discrete selection, it says: effects exist on a continuum from negligible to substantial, and we should shrink them proportionally to their signal strength. Small effects get heavily shrunk (possibly to near-zero), while large effects escape shrinkage almost entirely.\n",
    "\n",
    "$$ \\beta_{j} = \\tau \\cdot \\lambda_{j} \\cdot \\beta_{j\\text{raw}}$$\n",
    "\n",
    "where $\\tau$ is a global shrinkage parameter shared across all coefficients, and $\\lambda_{j}$ is local or specific to each coefficient and regularised so as to ensure finite variance. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "806df6ea",
   "metadata": {},
   "source": [
    "### Hyperparameters for Variable Selection Priors\n",
    "\n",
    "You can control the behaviour of the variable selection priors through some of the hyperparameters available. For the spike and slab prior, the most important hyperparamers are `temperature`, `pi_alpha`, and `pi_beta`. \n",
    "\n",
    "Because our sampler doesn't like discrete variables, we're approximating a bernoulli outcome in our sampling to define the spike and slab. The approximation is governed by the `temperature` parameter. The default value of 0.1 works well in most cases, creating indicators that cluster near 0 or 1 without causing sampling difficulties.\n",
    "\n",
    "The selection probability parameters `pi_alpha` and `pi_beta` encode your prior belief about sparsity. With both set to 2 (the default), you're placing a Beta(2,2) prior on π, the overall proportion of selected variables. This is symmetric around 0.5 but slightly concentrated there—you're saying \"I don't know how many variables are relevant, but probably not all of them and probably not none of them.\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ae848fe9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 3, figsize=(20, 6))\n",
    "axs = axs.flatten()\n",
    "axs[0].hist(pm.draw(pm.Beta.dist(2, 2), 1000), ec=\"black\", color=\"slateblue\")\n",
    "axs[1].hist(pm.draw(pm.Beta.dist(2, 5), 1000), ec=\"black\", color=\"slateblue\")\n",
    "axs[2].hist(pm.draw(pm.Beta.dist(5, 2), 1000), ec=\"black\", color=\"slateblue\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3237bb49",
   "metadata": {},
   "source": [
    "We'll now fit two models and estimate the implied treatment effect."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "763ca253",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--------------------------------------------------------------------------------\n",
      "Model 1: Normal Priors (No Variable Selection)\n",
      "--------------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/causalpy/experiments/instrumental_variable.py:187: UserWarning: Warning. The treatment variable is not Binary.\n",
      "                This is not necessarily a problem but it violates\n",
      "                the assumption of a simple IV experiment.\n",
      "                The coefficients should be interpreted appropriately.\n",
      "  warnings.warn(\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (4 chains in 4 jobs)\n",
      "NUTS: [beta_t, beta_z, chol_cov]\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b332c2837f1849329b3b561a9765f3e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains for 2_000 tune and 1_000 draw iterations (8_000 + 4_000 draws total) took 133 seconds.\n",
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/causalpy/experiments/instrumental_variable.py:187: UserWarning: Warning. The treatment variable is not Binary.\n",
      "                This is not necessarily a problem but it violates\n",
      "                the assumption of a simple IV experiment.\n",
      "                The coefficients should be interpreted appropriately.\n",
      "  warnings.warn(\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/causalpy/pymc_models.py:699: UserWarning: Variable selection priors specified. The 'mus' and 'sigmas' in the priors dict will be ignored for beta coefficients. Only 'eta' and 'lkj_sd' will be used.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--------------------------------------------------------------------------------\n",
      "Model 2: Spike-and-Slab Priors\n",
      "--------------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (4 chains in 4 jobs)\n",
      "NUTS: [pi_beta_t, beta_t_raw, gamma_beta_t_u, pi_beta_z, beta_z_raw, gamma_beta_z_u, chol_cov]\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db312f02efe743c386a4f2b4449f8904",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains for 2_000 tune and 1_000 draw iterations (8_000 + 4_000 draws total) took 551 seconds.\n",
      "There were 167 divergences after tuning. Increase `target_accept` or reparameterize.\n",
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n",
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    }
   ],
   "source": [
    "sample_kwargs = {\n",
    "    \"draws\": 1000,\n",
    "    \"tune\": 2000,\n",
    "    \"chains\": 4,\n",
    "    \"cores\": 4,\n",
    "    \"target_accept\": 0.95,\n",
    "    \"progressbar\": True,\n",
    "    \"random_seed\": 42,\n",
    "    \"mp_ctx\": \"spawn\",\n",
    "}\n",
    "\n",
    "# =========================================================================\n",
    "# Model 1: Normal priors (no selection)\n",
    "# =========================================================================\n",
    "print(\"\\n\" + \"-\" * 80)\n",
    "print(\"Model 1: Normal Priors (No Variable Selection)\")\n",
    "print(\"-\" * 80)\n",
    "\n",
    "result_normal = cp.InstrumentalVariable(\n",
    "    instruments_data=instruments_data,\n",
    "    data=data,\n",
    "    instruments_formula=instruments_formula,\n",
    "    formula=formula,\n",
    "    model=cp.pymc_models.InstrumentalVariableRegression(sample_kwargs=sample_kwargs),\n",
    "    vs_prior_type=None,  # No variable selection\n",
    ")\n",
    "\n",
    "# =========================================================================\n",
    "# Model 2: Spike-and-Slab priors\n",
    "# =========================================================================\n",
    "print(\"\\n\" + \"-\" * 80)\n",
    "print(\"Model 2: Spike-and-Slab Priors\")\n",
    "print(\"-\" * 80)\n",
    "\n",
    "result_spike_slab = cp.InstrumentalVariable(\n",
    "    instruments_data=instruments_data,\n",
    "    data=data,\n",
    "    instruments_formula=instruments_formula,\n",
    "    formula=formula,\n",
    "    model=cp.pymc_models.InstrumentalVariableRegression(sample_kwargs=sample_kwargs),\n",
    "    vs_prior_type=\"spike_and_slab\",\n",
    "    vs_hyperparams={\n",
    "        \"pi_alpha\": 2,\n",
    "        \"pi_beta\": 2,\n",
    "        \"slab_sigma\": 2,\n",
    "    },\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ccb6e0b",
   "metadata": {},
   "source": [
    "The models have quite a distinct structure. Compare the normal IV model with non variable selection priors. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e97a9ca2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 13.1.2 (0)\n",
       " -->\n",
       "<!-- Pages: 1 -->\n",
       "<svg width=\"566pt\" height=\"503pt\"\n",
       " viewBox=\"0.00 0.00 566.00 503.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 499.45)\">\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-499.45 562,-499.45 562,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\">\n",
       "<title>clusterinstruments (15)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M42,-365.63C42,-365.63 124,-365.63 124,-365.63 130,-365.63 136,-371.63 136,-377.63 136,-377.63 136,-475.45 136,-475.45 136,-481.45 130,-487.45 124,-487.45 124,-487.45 42,-487.45 42,-487.45 36,-487.45 30,-481.45 30,-475.45 30,-475.45 30,-377.63 30,-377.63 30,-371.63 36,-365.63 42,-365.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"83\" y=\"-372.83\" font-family=\"Times,serif\" font-size=\"14.00\">instruments (15)</text>\n",
       "</g>\n",
       "<g id=\"clust2\" class=\"cluster\">\n",
       "<title>clustercovariates (16)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M157,-365.63C157,-365.63 231,-365.63 231,-365.63 237,-365.63 243,-371.63 243,-377.63 243,-377.63 243,-475.45 243,-475.45 243,-481.45 237,-487.45 231,-487.45 231,-487.45 157,-487.45 157,-487.45 151,-487.45 145,-481.45 145,-475.45 145,-475.45 145,-377.63 145,-377.63 145,-371.63 151,-365.63 157,-365.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194.5\" y=\"-372.83\" font-family=\"Times,serif\" font-size=\"14.00\">covariates (16)</text>\n",
       "</g>\n",
       "<g id=\"clust3\" class=\"cluster\">\n",
       "<title>cluster3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M261,-129.82C261,-129.82 421,-129.82 421,-129.82 427,-129.82 433,-135.82 433,-141.82 433,-141.82 433,-239.63 433,-239.63 433,-245.63 427,-251.63 421,-251.63 421,-251.63 261,-251.63 261,-251.63 255,-251.63 249,-245.63 249,-239.63 249,-239.63 249,-141.82 249,-141.82 249,-135.82 255,-129.82 261,-129.82\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"421.62\" y=\"-137.02\" font-family=\"Times,serif\" font-size=\"14.00\">3</text>\n",
       "</g>\n",
       "<g id=\"clust4\" class=\"cluster\">\n",
       "<title>cluster2 x 2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M226,-19.91C226,-19.91 420,-19.91 420,-19.91 426,-19.91 432,-25.91 432,-31.91 432,-31.91 432,-105.91 432,-105.91 432,-111.91 426,-117.91 420,-117.91 420,-117.91 226,-117.91 226,-117.91 220,-117.91 214,-111.91 214,-105.91 214,-105.91 214,-31.91 214,-31.91 214,-25.91 220,-19.91 226,-19.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"410.12\" y=\"-27.11\" font-family=\"Times,serif\" font-size=\"14.00\">2 x 2</text>\n",
       "</g>\n",
       "<g id=\"clust5\" class=\"cluster\">\n",
       "<title>cluster2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M452,-19.91C452,-19.91 538,-19.91 538,-19.91 544,-19.91 550,-25.91 550,-31.91 550,-31.91 550,-105.91 550,-105.91 550,-111.91 544,-117.91 538,-117.91 538,-117.91 452,-117.91 452,-117.91 446,-117.91 440,-111.91 440,-105.91 440,-105.91 440,-31.91 440,-31.91 440,-25.91 446,-19.91 452,-19.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"538.62\" y=\"-27.11\" font-family=\"Times,serif\" font-size=\"14.00\">2</text>\n",
       "</g>\n",
       "<g id=\"clust6\" class=\"cluster\">\n",
       "<title>cluster2500</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M45,-259.63C45,-259.63 235,-259.63 235,-259.63 241,-259.63 247,-265.63 247,-271.63 247,-271.63 247,-345.63 247,-345.63 247,-351.63 241,-357.63 235,-357.63 235,-357.63 45,-357.63 45,-357.63 39,-357.63 33,-351.63 33,-345.63 33,-345.63 33,-271.63 33,-271.63 33,-265.63 39,-259.63 45,-259.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"225.5\" y=\"-266.83\" font-family=\"Times,serif\" font-size=\"14.00\">2500</text>\n",
       "</g>\n",
       "<g id=\"clust7\" class=\"cluster\">\n",
       "<title>cluster2500 x 2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M20,-8C20,-8 194,-8 194,-8 200,-8 206,-14 206,-20 206,-20 206,-227.73 206,-227.73 206,-233.73 200,-239.73 194,-239.73 194,-239.73 20,-239.73 20,-239.73 14,-239.73 8,-233.73 8,-227.73 8,-227.73 8,-20 8,-20 8,-14 14,-8 20,-8\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"174\" y=\"-15.2\" font-family=\"Times,serif\" font-size=\"14.00\">2500 x 2</text>\n",
       "</g>\n",
       "<!-- beta_t -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>beta_t</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"86\" cy=\"-438.79\" rx=\"41.01\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"86\" y=\"-450.24\" font-family=\"Times,serif\" font-size=\"14.00\">beta_t</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"86\" y=\"-433.74\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"86\" y=\"-417.24\" font-family=\"Times,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- mu_t -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>mu_t</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"131.12,-349.63 40.88,-349.63 40.88,-292.13 131.12,-292.13 131.12,-349.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"86\" y=\"-332.33\" font-family=\"Times,serif\" font-size=\"14.00\">mu_t</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"86\" y=\"-315.83\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"86\" y=\"-299.33\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- beta_t&#45;&gt;mu_t -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>beta_t&#45;&gt;mu_t</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M86,-397.9C86,-386.02 86,-373.01 86,-361.16\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"89.5,-361.3 86,-351.3 82.5,-361.3 89.5,-361.3\"/>\n",
       "</g>\n",
       "<!-- beta_z -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>beta_z</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"194\" cy=\"-438.79\" rx=\"41.01\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194\" y=\"-450.24\" font-family=\"Times,serif\" font-size=\"14.00\">beta_z</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194\" y=\"-433.74\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194\" y=\"-417.24\" font-family=\"Times,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- mu_y -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>mu_y</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"239.12,-349.63 148.88,-349.63 148.88,-292.13 239.12,-292.13 239.12,-349.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194\" y=\"-332.33\" font-family=\"Times,serif\" font-size=\"14.00\">mu_y</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194\" y=\"-315.83\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194\" y=\"-299.33\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- beta_z&#45;&gt;mu_y -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>beta_z&#45;&gt;mu_y</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M194,-397.9C194,-386.02 194,-373.01 194,-361.16\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"197.5,-361.3 194,-351.3 190.5,-361.3 197.5,-361.3\"/>\n",
       "</g>\n",
       "<!-- chol_cov -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>chol_cov</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"341\" cy=\"-202.98\" rx=\"84.5\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"341\" y=\"-214.43\" font-family=\"Times,serif\" font-size=\"14.00\">chol_cov</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"341\" y=\"-197.93\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"341\" y=\"-181.43\" font-family=\"Times,serif\" font-size=\"14.00\">_LKJCholeskyCov</text>\n",
       "</g>\n",
       "<!-- cov -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>cov</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"312.12,-109.91 221.88,-109.91 221.88,-52.41 312.12,-52.41 312.12,-109.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"267\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">cov</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"267\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"267\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;cov -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;cov</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M317.38,-163.74C308.74,-149.74 298.97,-133.92 290.34,-119.96\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"293.33,-118.13 285.1,-111.46 287.37,-121.81 293.33,-118.13\"/>\n",
       "</g>\n",
       "<!-- chol_cov_corr -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>chol_cov_corr</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"424,-109.91 330,-109.91 330,-52.41 424,-52.41 424,-109.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"377\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">chol_cov_corr</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"377\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"377\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;chol_cov_corr -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;chol_cov_corr</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M352.89,-162.41C356.89,-149.1 361.34,-134.26 365.33,-121.01\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"368.58,-122.34 368.11,-111.76 361.88,-120.33 368.58,-122.34\"/>\n",
       "</g>\n",
       "<!-- chol_cov_stds -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>chol_cov_stds</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"541.62,-109.91 448.38,-109.91 448.38,-52.41 541.62,-52.41 541.62,-109.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"495\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">chol_cov_stds</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"495\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"495\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;chol_cov_stds -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;chol_cov_stds</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M384.72,-167.96C405.09,-152.11 429.31,-133.27 449.86,-117.28\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"451.78,-120.22 457.52,-111.32 447.48,-114.7 451.78,-120.22\"/>\n",
       "</g>\n",
       "<!-- likelihood -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>likelihood</title>\n",
       "<ellipse fill=\"lightgrey\" stroke=\"black\" cx=\"107\" cy=\"-81.16\" rx=\"91.39\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"107\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">likelihood</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"107\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"107\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Multivariate_normal</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;likelihood -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M287.28,-171.27C263.76,-158.12 235.72,-142.83 210,-129.82 200.61,-125.06 190.64,-120.22 180.76,-115.53\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"182.43,-112.44 171.89,-111.35 179.45,-118.77 182.43,-112.44\"/>\n",
       "</g>\n",
       "<!-- mu -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>mu</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"152.12,-231.73 61.88,-231.73 61.88,-174.23 152.12,-174.23 152.12,-231.73\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"107\" y=\"-214.43\" font-family=\"Times,serif\" font-size=\"14.00\">mu</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"107\" y=\"-197.93\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"107\" y=\"-181.43\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- mu_y&#45;&gt;mu -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>mu_y&#45;&gt;mu</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M172.94,-291.83C161.48,-276.56 147.17,-257.49 134.8,-241.02\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"137.88,-239.28 129.08,-233.39 132.28,-243.49 137.88,-239.28\"/>\n",
       "</g>\n",
       "<!-- mu_t&#45;&gt;mu -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>mu_t&#45;&gt;mu</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M91.08,-291.83C93.72,-277.27 96.99,-259.25 99.87,-243.32\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"103.28,-244.13 101.62,-233.67 96.39,-242.88 103.28,-244.13\"/>\n",
       "</g>\n",
       "<!-- mu&#45;&gt;likelihood -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>mu&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M107,-173.88C107,-161.78 107,-147.23 107,-133.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"110.5,-133.38 107,-123.38 103.5,-133.38 110.5,-133.38\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x28eaab230>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.model_to_graphviz(result_normal.model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34f3a1b7",
   "metadata": {},
   "source": [
    "Now compare the structure of the spike and slab model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4f8c2685",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 13.1.2 (0)\n",
       " -->\n",
       "<!-- Pages: 1 -->\n",
       "<svg width=\"811pt\" height=\"714pt\"\n",
       " viewBox=\"0.00 0.00 811.00 714.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 710.27)\">\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-710.27 807.44,-710.27 807.44,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\">\n",
       "<title>clusterinstruments (15)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M369.44,-365.63C369.44,-365.63 584.44,-365.63 584.44,-365.63 590.44,-365.63 596.44,-371.63 596.44,-377.63 596.44,-377.63 596.44,-686.27 596.44,-686.27 596.44,-692.27 590.44,-698.27 584.44,-698.27 584.44,-698.27 369.44,-698.27 369.44,-698.27 363.44,-698.27 357.44,-692.27 357.44,-686.27 357.44,-686.27 357.44,-377.63 357.44,-377.63 357.44,-371.63 363.44,-365.63 369.44,-365.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"543.44\" y=\"-372.83\" font-family=\"Times,serif\" font-size=\"14.00\">instruments (15)</text>\n",
       "</g>\n",
       "<g id=\"clust2\" class=\"cluster\">\n",
       "<title>clustercovariates (16)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M117.44,-365.63C117.44,-365.63 337.44,-365.63 337.44,-365.63 343.44,-365.63 349.44,-371.63 349.44,-377.63 349.44,-377.63 349.44,-686.27 349.44,-686.27 349.44,-692.27 343.44,-698.27 337.44,-698.27 337.44,-698.27 117.44,-698.27 117.44,-698.27 111.44,-698.27 105.44,-692.27 105.44,-686.27 105.44,-686.27 105.44,-377.63 105.44,-377.63 105.44,-371.63 111.44,-365.63 117.44,-365.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"300.94\" y=\"-372.83\" font-family=\"Times,serif\" font-size=\"14.00\">covariates (16)</text>\n",
       "</g>\n",
       "<g id=\"clust3\" class=\"cluster\">\n",
       "<title>cluster3</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M506.44,-129.82C506.44,-129.82 666.44,-129.82 666.44,-129.82 672.44,-129.82 678.44,-135.82 678.44,-141.82 678.44,-141.82 678.44,-239.63 678.44,-239.63 678.44,-245.63 672.44,-251.63 666.44,-251.63 666.44,-251.63 506.44,-251.63 506.44,-251.63 500.44,-251.63 494.44,-245.63 494.44,-239.63 494.44,-239.63 494.44,-141.82 494.44,-141.82 494.44,-135.82 500.44,-129.82 506.44,-129.82\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"667.06\" y=\"-137.02\" font-family=\"Times,serif\" font-size=\"14.00\">3</text>\n",
       "</g>\n",
       "<g id=\"clust4\" class=\"cluster\">\n",
       "<title>cluster2 x 2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M471.44,-19.91C471.44,-19.91 665.44,-19.91 665.44,-19.91 671.44,-19.91 677.44,-25.91 677.44,-31.91 677.44,-31.91 677.44,-105.91 677.44,-105.91 677.44,-111.91 671.44,-117.91 665.44,-117.91 665.44,-117.91 471.44,-117.91 471.44,-117.91 465.44,-117.91 459.44,-111.91 459.44,-105.91 459.44,-105.91 459.44,-31.91 459.44,-31.91 459.44,-25.91 465.44,-19.91 471.44,-19.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"655.56\" y=\"-27.11\" font-family=\"Times,serif\" font-size=\"14.00\">2 x 2</text>\n",
       "</g>\n",
       "<g id=\"clust5\" class=\"cluster\">\n",
       "<title>cluster2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M697.44,-19.91C697.44,-19.91 783.44,-19.91 783.44,-19.91 789.44,-19.91 795.44,-25.91 795.44,-31.91 795.44,-31.91 795.44,-105.91 795.44,-105.91 795.44,-111.91 789.44,-117.91 783.44,-117.91 783.44,-117.91 697.44,-117.91 697.44,-117.91 691.44,-117.91 685.44,-111.91 685.44,-105.91 685.44,-105.91 685.44,-31.91 685.44,-31.91 685.44,-25.91 691.44,-19.91 697.44,-19.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"784.06\" y=\"-27.11\" font-family=\"Times,serif\" font-size=\"14.00\">2</text>\n",
       "</g>\n",
       "<g id=\"clust6\" class=\"cluster\">\n",
       "<title>cluster2500</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M257.44,-259.63C257.44,-259.63 447.44,-259.63 447.44,-259.63 453.44,-259.63 459.44,-265.63 459.44,-271.63 459.44,-271.63 459.44,-345.63 459.44,-345.63 459.44,-351.63 453.44,-357.63 447.44,-357.63 447.44,-357.63 257.44,-357.63 257.44,-357.63 251.44,-357.63 245.44,-351.63 245.44,-345.63 245.44,-345.63 245.44,-271.63 245.44,-271.63 245.44,-265.63 251.44,-259.63 257.44,-259.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"437.94\" y=\"-266.83\" font-family=\"Times,serif\" font-size=\"14.00\">2500</text>\n",
       "</g>\n",
       "<g id=\"clust7\" class=\"cluster\">\n",
       "<title>cluster2500 x 2</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M265.44,-8C265.44,-8 439.44,-8 439.44,-8 445.44,-8 451.44,-14 451.44,-20 451.44,-20 451.44,-227.73 451.44,-227.73 451.44,-233.73 445.44,-239.73 439.44,-239.73 439.44,-239.73 265.44,-239.73 265.44,-239.73 259.44,-239.73 253.44,-233.73 253.44,-227.73 253.44,-227.73 253.44,-20 253.44,-20 253.44,-14 259.44,-8 265.44,-8\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-15.2\" font-family=\"Times,serif\" font-size=\"14.00\">2500 x 2</text>\n",
       "</g>\n",
       "<!-- pi_beta_z -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>pi_beta_z</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"48.44\" cy=\"-649.61\" rx=\"48.44\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"48.44\" y=\"-661.06\" font-family=\"Times,serif\" font-size=\"14.00\">pi_beta_z</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"48.44\" y=\"-644.56\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"48.44\" y=\"-628.06\" font-family=\"Times,serif\" font-size=\"14.00\">Beta</text>\n",
       "</g>\n",
       "<!-- gamma_beta_z -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>gamma_beta_z</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"213.06,-561.04 113.81,-561.04 113.81,-503.54 213.06,-503.54 213.06,-561.04\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"163.44\" y=\"-543.74\" font-family=\"Times,serif\" font-size=\"14.00\">gamma_beta_z</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"163.44\" y=\"-527.24\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"163.44\" y=\"-510.74\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- pi_beta_z&#45;&gt;gamma_beta_z -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>pi_beta_z&#45;&gt;gamma_beta_z</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M79.25,-617.71C94.1,-602.82 111.99,-584.88 127.46,-569.37\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"129.64,-572.14 134.22,-562.59 124.68,-567.2 129.64,-572.14\"/>\n",
       "</g>\n",
       "<!-- pi_beta_t -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>pi_beta_t</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"652.44\" cy=\"-649.61\" rx=\"46.85\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"652.44\" y=\"-661.06\" font-family=\"Times,serif\" font-size=\"14.00\">pi_beta_t</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"652.44\" y=\"-644.56\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"652.44\" y=\"-628.06\" font-family=\"Times,serif\" font-size=\"14.00\">Beta</text>\n",
       "</g>\n",
       "<!-- gamma_beta_t -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>gamma_beta_t</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"587.94,-561.04 490.94,-561.04 490.94,-503.54 587.94,-503.54 587.94,-561.04\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"539.44\" y=\"-543.74\" font-family=\"Times,serif\" font-size=\"14.00\">gamma_beta_t</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"539.44\" y=\"-527.24\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"539.44\" y=\"-510.74\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- pi_beta_t&#45;&gt;gamma_beta_t -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>pi_beta_t&#45;&gt;gamma_beta_t</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M622.45,-618.01C607.75,-603.01 589.96,-584.85 574.62,-569.2\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"577.41,-567.04 567.91,-562.35 572.41,-571.94 577.41,-567.04\"/>\n",
       "</g>\n",
       "<!-- beta_t_raw -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>beta_t_raw</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"419.44\" cy=\"-532.29\" rx=\"53.74\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-543.74\" font-family=\"Times,serif\" font-size=\"14.00\">beta_t_raw</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-527.24\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-510.74\" font-family=\"Times,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- beta_t -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>beta_t</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"464.56,-455.63 374.31,-455.63 374.31,-398.13 464.56,-398.13 464.56,-455.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-438.33\" font-family=\"Times,serif\" font-size=\"14.00\">beta_t</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-421.83\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"419.44\" y=\"-405.33\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- beta_t_raw&#45;&gt;beta_t -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>beta_t_raw&#45;&gt;beta_t</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M419.44,-491.31C419.44,-483.46 419.44,-475.23 419.44,-467.4\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"422.94,-467.42 419.44,-457.42 415.94,-467.42 422.94,-467.42\"/>\n",
       "</g>\n",
       "<!-- mu_t -->\n",
       "<g id=\"node16\" class=\"node\">\n",
       "<title>mu_t</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"451.56,-349.63 361.31,-349.63 361.31,-292.13 451.56,-292.13 451.56,-349.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"406.44\" y=\"-332.33\" font-family=\"Times,serif\" font-size=\"14.00\">mu_t</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"406.44\" y=\"-315.83\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"406.44\" y=\"-299.33\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- beta_t&#45;&gt;mu_t -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>beta_t&#45;&gt;mu_t</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M415.92,-397.74C414.52,-386.53 412.88,-373.42 411.36,-361.29\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"414.86,-361.04 410.15,-351.55 407.91,-361.91 414.86,-361.04\"/>\n",
       "</g>\n",
       "<!-- gamma_beta_t&#45;&gt;beta_t -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>gamma_beta_t&#45;&gt;beta_t</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M506.96,-503.31C492.69,-491.01 475.76,-476.42 460.62,-463.37\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"462.94,-460.76 453.08,-456.88 458.37,-466.06 462.94,-460.76\"/>\n",
       "</g>\n",
       "<!-- gamma_beta_t_u -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>gamma_beta_t_u</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"509.44\" cy=\"-649.61\" rx=\"78.14\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"509.44\" y=\"-661.06\" font-family=\"Times,serif\" font-size=\"14.00\">gamma_beta_t_u</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"509.44\" y=\"-644.56\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"509.44\" y=\"-628.06\" font-family=\"Times,serif\" font-size=\"14.00\">Uniform</text>\n",
       "</g>\n",
       "<!-- gamma_beta_t_u&#45;&gt;gamma_beta_t -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>gamma_beta_t_u&#45;&gt;gamma_beta_t</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M519.76,-608.92C522.84,-597.08 526.22,-584.1 529.29,-572.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"532.61,-573.43 531.74,-562.88 525.84,-571.67 532.61,-573.43\"/>\n",
       "</g>\n",
       "<!-- gamma_beta_z_u -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>gamma_beta_z_u</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"194.44\" cy=\"-649.61\" rx=\"79.73\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194.44\" y=\"-661.06\" font-family=\"Times,serif\" font-size=\"14.00\">gamma_beta_z_u</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194.44\" y=\"-644.56\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"194.44\" y=\"-628.06\" font-family=\"Times,serif\" font-size=\"14.00\">Uniform</text>\n",
       "</g>\n",
       "<!-- gamma_beta_z_u&#45;&gt;gamma_beta_z -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>gamma_beta_z_u&#45;&gt;gamma_beta_z</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M183.77,-608.92C180.58,-597.08 177.1,-584.1 173.92,-572.3\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"177.36,-571.62 171.39,-562.87 170.6,-573.44 177.36,-571.62\"/>\n",
       "</g>\n",
       "<!-- beta_z -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>beta_z</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"331.56,-455.63 241.31,-455.63 241.31,-398.13 331.56,-398.13 331.56,-455.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"286.44\" y=\"-438.33\" font-family=\"Times,serif\" font-size=\"14.00\">beta_z</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"286.44\" y=\"-421.83\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"286.44\" y=\"-405.33\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- mu_y -->\n",
       "<g id=\"node15\" class=\"node\">\n",
       "<title>mu_y</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"343.56,-349.63 253.31,-349.63 253.31,-292.13 343.56,-292.13 343.56,-349.63\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"298.44\" y=\"-332.33\" font-family=\"Times,serif\" font-size=\"14.00\">mu_y</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"298.44\" y=\"-315.83\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"298.44\" y=\"-299.33\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- beta_z&#45;&gt;mu_y -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>beta_z&#45;&gt;mu_y</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M289.68,-397.74C290.98,-386.53 292.49,-373.42 293.89,-361.29\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"297.34,-361.89 295.01,-351.55 290.39,-361.09 297.34,-361.89\"/>\n",
       "</g>\n",
       "<!-- gamma_beta_z&#45;&gt;beta_z -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>gamma_beta_z&#45;&gt;beta_z</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M196.72,-503.31C211.49,-490.89 229.03,-476.15 244.66,-463\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"246.56,-465.98 251.97,-456.86 242.06,-460.62 246.56,-465.98\"/>\n",
       "</g>\n",
       "<!-- beta_z_raw -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>beta_z_raw</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"286.44\" cy=\"-532.29\" rx=\"55.33\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"286.44\" y=\"-543.74\" font-family=\"Times,serif\" font-size=\"14.00\">beta_z_raw</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"286.44\" y=\"-527.24\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"286.44\" y=\"-510.74\" font-family=\"Times,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- beta_z_raw&#45;&gt;beta_z -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>beta_z_raw&#45;&gt;beta_z</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M286.44,-491.31C286.44,-483.46 286.44,-475.23 286.44,-467.4\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"289.94,-467.42 286.44,-457.42 282.94,-467.42 289.94,-467.42\"/>\n",
       "</g>\n",
       "<!-- chol_cov -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>chol_cov</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"586.44\" cy=\"-202.98\" rx=\"84.5\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"586.44\" y=\"-214.43\" font-family=\"Times,serif\" font-size=\"14.00\">chol_cov</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"586.44\" y=\"-197.93\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"586.44\" y=\"-181.43\" font-family=\"Times,serif\" font-size=\"14.00\">_LKJCholeskyCov</text>\n",
       "</g>\n",
       "<!-- cov -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>cov</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"557.56,-109.91 467.31,-109.91 467.31,-52.41 557.56,-52.41 557.56,-109.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"512.44\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">cov</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"512.44\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"512.44\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;cov -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;cov</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M562.82,-163.74C554.17,-149.74 544.4,-133.92 535.78,-119.96\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"538.77,-118.13 530.53,-111.46 532.81,-121.81 538.77,-118.13\"/>\n",
       "</g>\n",
       "<!-- chol_cov_corr -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>chol_cov_corr</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"669.44,-109.91 575.44,-109.91 575.44,-52.41 669.44,-52.41 669.44,-109.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"622.44\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">chol_cov_corr</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"622.44\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"622.44\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;chol_cov_corr -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;chol_cov_corr</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M598.32,-162.41C602.32,-149.1 606.78,-134.26 610.76,-121.01\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"614.02,-122.34 613.54,-111.76 607.31,-120.33 614.02,-122.34\"/>\n",
       "</g>\n",
       "<!-- chol_cov_stds -->\n",
       "<g id=\"node14\" class=\"node\">\n",
       "<title>chol_cov_stds</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"787.06,-109.91 693.81,-109.91 693.81,-52.41 787.06,-52.41 787.06,-109.91\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"740.44\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">chol_cov_stds</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"740.44\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"740.44\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;chol_cov_stds -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;chol_cov_stds</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M630.16,-167.96C650.53,-152.11 674.75,-133.27 695.3,-117.28\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"697.21,-120.22 702.96,-111.32 692.91,-114.7 697.21,-120.22\"/>\n",
       "</g>\n",
       "<!-- likelihood -->\n",
       "<g id=\"node18\" class=\"node\">\n",
       "<title>likelihood</title>\n",
       "<ellipse fill=\"lightgrey\" stroke=\"black\" cx=\"352.44\" cy=\"-81.16\" rx=\"91.39\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"352.44\" y=\"-92.61\" font-family=\"Times,serif\" font-size=\"14.00\">likelihood</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"352.44\" y=\"-76.11\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"352.44\" y=\"-59.61\" font-family=\"Times,serif\" font-size=\"14.00\">Multivariate_normal</text>\n",
       "</g>\n",
       "<!-- chol_cov&#45;&gt;likelihood -->\n",
       "<g id=\"edge17\" class=\"edge\">\n",
       "<title>chol_cov&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M532.72,-171.27C509.2,-158.12 481.15,-142.83 455.44,-129.82 446.04,-125.06 436.08,-120.22 426.2,-115.53\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"427.87,-112.44 417.33,-111.35 424.88,-118.77 427.87,-112.44\"/>\n",
       "</g>\n",
       "<!-- mu -->\n",
       "<g id=\"node17\" class=\"node\">\n",
       "<title>mu</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"397.56,-231.73 307.31,-231.73 307.31,-174.23 397.56,-174.23 397.56,-231.73\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"352.44\" y=\"-214.43\" font-family=\"Times,serif\" font-size=\"14.00\">mu</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"352.44\" y=\"-197.93\" font-family=\"Times,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"352.44\" y=\"-181.43\" font-family=\"Times,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- mu_y&#45;&gt;mu -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>mu_y&#45;&gt;mu</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M311.51,-291.83C318.42,-276.99 327.01,-258.55 334.54,-242.4\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"337.61,-244.09 338.66,-233.55 331.26,-241.14 337.61,-244.09\"/>\n",
       "</g>\n",
       "<!-- mu_t&#45;&gt;mu -->\n",
       "<g id=\"edge15\" class=\"edge\">\n",
       "<title>mu_t&#45;&gt;mu</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.36,-291.83C386.45,-276.99 377.86,-258.55 370.34,-242.4\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"373.61,-241.14 366.21,-233.55 367.27,-244.09 373.61,-241.14\"/>\n",
       "</g>\n",
       "<!-- mu&#45;&gt;likelihood -->\n",
       "<g id=\"edge16\" class=\"edge\">\n",
       "<title>mu&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M352.44,-173.88C352.44,-161.78 352.44,-147.23 352.44,-133.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"355.94,-133.38 352.44,-123.38 348.94,-133.38 355.94,-133.38\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x2f1d7ac50>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm.model_to_graphviz(result_spike_slab.model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "368660c8",
   "metadata": {},
   "source": [
    "Despite seeing some divergences in our spike and slab model, most other sampler health metrics seem healthy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "0755095c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "az.plot_energy(result_spike_slab.idata, figsize=(20, 6));"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bffd8b6",
   "metadata": {},
   "source": [
    "And since we know the true data generating conditions we can also assess the derived posterior treatment estimates. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "838e0726",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(20, 6))\n",
    "az.plot_posterior(\n",
    "    result_normal.idata,\n",
    "    var_names=[\"beta_z\"],\n",
    "    coords={\"covariates\": [\"T_cont\"]},\n",
    "    ax=ax,\n",
    "    label=\"Normal\",\n",
    ")\n",
    "az.plot_posterior(\n",
    "    result_spike_slab.idata,\n",
    "    var_names=[\"beta_z\"],\n",
    "    coords={\"covariates\": [\"T_cont\"]},\n",
    "    ax=ax,\n",
    "    color=\"green\",\n",
    "    label=\"spike and slab\",\n",
    ")\n",
    "ax.axvline(3, color=\"black\", linestyle=\"--\", label=\"True value\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "057b4f5d",
   "metadata": {},
   "source": [
    "This plot suggests that the spike and slab prior was better able to ignore noise in the process and zero in on the true effect. This will not always work but it is a sensible practice to at least sensitivity check difference between the estimates under different prior settings. We can observe how aggressively the spike and slab prior worked to cull unwanted variables from each model by comparing the values on the coefficients across each model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "127888b7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x820 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "axs = az.plot_forest(\n",
    "    [result_spike_slab.idata, result_normal.idata],\n",
    "    var_names=[\"beta_z\"],\n",
    "    combined=True,\n",
    "    model_names=[\"Spike and Slab\", \"Normal\"],\n",
    "    r_hat=True,\n",
    ")\n",
    "axs[0].set_title(\"Parameter Comparison Outcome Model \\n Baseline v Spike and Slab\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f09b24bf",
   "metadata": {},
   "source": [
    "#### The Treatment Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "acafc928",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "axs = az.plot_forest(\n",
    "    [result_spike_slab.idata, result_normal.idata],\n",
    "    var_names=[\"beta_t\"],\n",
    "    combined=True,\n",
    "    model_names=[\"Spike and Slab\", \"Normal\"],\n",
    "    r_hat=True,\n",
    ")\n",
    "\n",
    "axs[0].set_title(\"Parameter Comparison Treatment Model \\n Baseline v Spike and Slab\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f2a0b213",
   "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>prob</th>\n",
       "      <th>selected</th>\n",
       "      <th>gamma_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.01750</td>\n",
       "      <td>False</td>\n",
       "      <td>0.020546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.00000</td>\n",
       "      <td>True</td>\n",
       "      <td>0.991285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.64100</td>\n",
       "      <td>True</td>\n",
       "      <td>0.659357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.58200</td>\n",
       "      <td>True</td>\n",
       "      <td>0.616580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.17625</td>\n",
       "      <td>False</td>\n",
       "      <td>0.192949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.06075</td>\n",
       "      <td>False</td>\n",
       "      <td>0.068297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.66600</td>\n",
       "      <td>True</td>\n",
       "      <td>0.702671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.01000</td>\n",
       "      <td>False</td>\n",
       "      <td>0.012679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.01300</td>\n",
       "      <td>False</td>\n",
       "      <td>0.016342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.00975</td>\n",
       "      <td>False</td>\n",
       "      <td>0.012392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.01700</td>\n",
       "      <td>False</td>\n",
       "      <td>0.019508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.01625</td>\n",
       "      <td>False</td>\n",
       "      <td>0.021217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.02000</td>\n",
       "      <td>False</td>\n",
       "      <td>0.024469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.01825</td>\n",
       "      <td>False</td>\n",
       "      <td>0.023474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.02700</td>\n",
       "      <td>False</td>\n",
       "      <td>0.031636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.01650</td>\n",
       "      <td>False</td>\n",
       "      <td>0.020863</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       prob  selected  gamma_mean\n",
       "0   0.01750     False    0.020546\n",
       "1   1.00000      True    0.991285\n",
       "2   0.64100      True    0.659357\n",
       "3   0.58200      True    0.616580\n",
       "4   0.17625     False    0.192949\n",
       "5   0.06075     False    0.068297\n",
       "6   0.66600      True    0.702671\n",
       "7   0.01000     False    0.012679\n",
       "8   0.01300     False    0.016342\n",
       "9   0.00975     False    0.012392\n",
       "10  0.01700     False    0.019508\n",
       "11  0.01625     False    0.021217\n",
       "12  0.02000     False    0.024469\n",
       "13  0.01825     False    0.023474\n",
       "14  0.02700     False    0.031636\n",
       "15  0.01650     False    0.020863"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary = result_spike_slab.model.vs_prior_outcome.get_inclusion_probabilities(\n",
    "    result_spike_slab.idata, \"beta_z\"\n",
    ")\n",
    "summary"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38568d27",
   "metadata": {},
   "source": [
    "## Horseshoe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "63edfa4e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--------------------------------------------------------------------------------\n",
      "Model 3: Horseshoe Priors\n",
      "--------------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/causalpy/experiments/instrumental_variable.py:187: UserWarning: Warning. The treatment variable is not Binary.\n",
      "                This is not necessarily a problem but it violates\n",
      "                the assumption of a simple IV experiment.\n",
      "                The coefficients should be interpreted appropriately.\n",
      "  warnings.warn(\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/causalpy/pymc_models.py:699: UserWarning: Variable selection priors specified. The 'mus' and 'sigmas' in the priors dict will be ignored for beta coefficients. Only 'eta' and 'lkj_sd' will be used.\n",
      "  warnings.warn(\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (4 chains in 4 jobs)\n",
      "NUTS: [tau_beta_t, lambda_beta_t, c2_beta_t, beta_t_raw, tau_beta_z, lambda_beta_z, c2_beta_z, beta_z_raw, chol_cov]\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d795e2e885b74a708646d93e4d96f152",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n",
      "/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/pytensor/compile/function/types.py:1039: RuntimeWarning: invalid value encountered in accumulate\n",
      "  outputs = vm() if output_subset is None else vm(output_subset=output_subset)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains for 2_000 tune and 1_000 draw iterations (8_000 + 4_000 draws total) took 644 seconds.\n",
      "There were 11 divergences after tuning. Increase `target_accept` or reparameterize.\n",
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    }
   ],
   "source": [
    "# =========================================================================\n",
    "# Model 2: Horseshoe priors\n",
    "# =========================================================================\n",
    "print(\"\\n\" + \"-\" * 80)\n",
    "print(\"Model 3: Horseshoe Priors\")\n",
    "print(\"-\" * 80)\n",
    "\n",
    "result_horseshoe = cp.InstrumentalVariable(\n",
    "    instruments_data=instruments_data,\n",
    "    data=data,\n",
    "    instruments_formula=instruments_formula,\n",
    "    formula=formula,\n",
    "    model=cp.pymc_models.InstrumentalVariableRegression(sample_kwargs=sample_kwargs),\n",
    "    vs_prior_type=\"horseshoe\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9c283ee1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[autoreload of cutils_ext failed: Traceback (most recent call last):\n",
      "  File \"/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/IPython/extensions/autoreload.py\", line 325, in check\n",
      "    superreload(m, reload, self.old_objects)\n",
      "    ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
      "  File \"/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/site-packages/IPython/extensions/autoreload.py\", line 580, in superreload\n",
      "    module = reload(module)\n",
      "  File \"/Users/nathanielforde/mambaforge/envs/CausalPy/lib/python3.13/importlib/__init__.py\", line 128, in reload\n",
      "    raise ModuleNotFoundError(f\"spec not found for the module {name!r}\", name=name)\n",
      "ModuleNotFoundError: spec not found for the module 'cutils_ext'\n",
      "]\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>shrinkage_factor</th>\n",
       "      <th>lambda_tilde</th>\n",
       "      <th>tau</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.016681</td>\n",
       "      <td>1.222294</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.847573</td>\n",
       "      <td>62.105904</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.164625</td>\n",
       "      <td>12.062918</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.136282</td>\n",
       "      <td>9.986066</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.090871</td>\n",
       "      <td>6.658563</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.030253</td>\n",
       "      <td>2.216801</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.199524</td>\n",
       "      <td>14.620131</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.015415</td>\n",
       "      <td>1.129522</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.015330</td>\n",
       "      <td>1.123328</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.015992</td>\n",
       "      <td>1.171778</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.015676</td>\n",
       "      <td>1.148635</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.020536</td>\n",
       "      <td>1.504805</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.021961</td>\n",
       "      <td>1.609202</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.020631</td>\n",
       "      <td>1.511758</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.021207</td>\n",
       "      <td>1.553953</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.019708</td>\n",
       "      <td>1.444109</td>\n",
       "      <td>0.038199</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    shrinkage_factor  lambda_tilde       tau\n",
       "0           0.016681      1.222294  0.038199\n",
       "1           0.847573     62.105904  0.038199\n",
       "2           0.164625     12.062918  0.038199\n",
       "3           0.136282      9.986066  0.038199\n",
       "4           0.090871      6.658563  0.038199\n",
       "5           0.030253      2.216801  0.038199\n",
       "6           0.199524     14.620131  0.038199\n",
       "7           0.015415      1.129522  0.038199\n",
       "8           0.015330      1.123328  0.038199\n",
       "9           0.015992      1.171778  0.038199\n",
       "10          0.015676      1.148635  0.038199\n",
       "11          0.020536      1.504805  0.038199\n",
       "12          0.021961      1.609202  0.038199\n",
       "13          0.020631      1.511758  0.038199\n",
       "14          0.021207      1.553953  0.038199\n",
       "15          0.019708      1.444109  0.038199"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary = result_horseshoe.model.vs_prior_outcome.get_shrinkage_factors(\n",
    "    result_horseshoe.idata, \"beta_z\"\n",
    ")\n",
    "summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "82b0121c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x2a238b390>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(20, 6))\n",
    "az.plot_posterior(\n",
    "    result_normal.idata, var_names=[\"beta_z\"], coords={\"covariates\": [\"T_cont\"]}, ax=ax\n",
    ")\n",
    "az.plot_posterior(\n",
    "    result_horseshoe.idata,\n",
    "    var_names=[\"beta_z\"],\n",
    "    coords={\"covariates\": [\"T_cont\"]},\n",
    "    ax=ax,\n",
    ")\n",
    "ax.axvline(3, color=\"black\", linestyle=\"--\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "CausalPy",
   "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": "3.13.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}