{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple example on using Instrumental Variables method for estimation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import patsy as ps\n",
    "\n",
    "from statsmodels.sandbox.regression.gmm import IV2SLS\n",
    "import os, sys\n",
    "from dowhy import CausalModel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the dataset\n",
    "\n",
    "We create a fictitious dataset with the goal of estimating the impact of education on future earnings of an individual. The `ability`  of the individual is a confounder and being given an `education_voucher` is the instrument."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_points = 1000\n",
    "education_abilty = 1\n",
    "education_voucher = 2\n",
    "income_abilty = 2\n",
    "income_education = 4\n",
    "\n",
    "\n",
    "# confounder\n",
    "ability = np.random.normal(0, 3, size=n_points)\n",
    "\n",
    "# instrument\n",
    "voucher = np.random.normal(2, 1, size=n_points) \n",
    "\n",
    "# treatment\n",
    "education = np.random.normal(5, 1, size=n_points) + education_abilty * ability +\\\n",
    "            education_voucher * voucher\n",
    "\n",
    "# outcome\n",
    "income = np.random.normal(10, 3, size=n_points) +\\\n",
    "         income_abilty * ability + income_education * education\n",
    "\n",
    "# build dataset (exclude confounder `ability` which we assume to be unobserved)\n",
    "data = np.stack([education, income, voucher]).T\n",
    "df = pd.DataFrame(data, columns = ['education', 'income', 'voucher'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using DoWhy to estimate the causal effect of education on future income\n",
    "\n",
    "We follow the four steps: \n",
    "1) model the problem using causal graph, \n",
    "\n",
    "2) identify if the causal effect can be estimated from the observed variables, \n",
    "\n",
    "3) estimate the effect, and \n",
    "\n",
    "4) check the robustness of the estimate. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Step 1: Model\n",
    "model=CausalModel(\n",
    "        data = df,\n",
    "        treatment='education',\n",
    "        outcome='income',\n",
    "        common_causes=['U'],\n",
    "        instruments=['voucher']\n",
    "        )\n",
    "model.view_model()\n",
    "from IPython.display import Image, display\n",
    "display(Image(filename=\"causal_model.png\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimand type: nonparametric-ate\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 2\n",
      "Estimand name: iv\n",
      "Estimand expression:\n",
      "Expectation(Derivative(income, [voucher])*Derivative([education], [voucher])**\n",
      "(-1))\n",
      "Estimand assumption 1, As-if-random: If U→→income then ¬(U →→{voucher})\n",
      "Estimand assumption 2, Exclusion: If we remove {voucher}→{education}, then ¬({voucher}→income)\n",
      "\n",
      "### Estimand : 3\n",
      "Estimand name: frontdoor\n",
      "No such variable(s) found!\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Step 2: Identify\n",
    "identified_estimand = model.identify_effect(proceed_when_unidentifiable=True)\n",
    "print(identified_estimand)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** Causal Estimate ***\n",
      "\n",
      "## Identified estimand\n",
      "Estimand type: nonparametric-ate\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: iv\n",
      "Estimand expression:\n",
      "Expectation(Derivative(income, [voucher])*Derivative([education], [voucher])**\n",
      "(-1))\n",
      "Estimand assumption 1, As-if-random: If U→→income then ¬(U →→{voucher})\n",
      "Estimand assumption 2, Exclusion: If we remove {voucher}→{education}, then ¬({voucher}→income)\n",
      "\n",
      "## Realized estimand\n",
      "Realized estimand: Wald Estimator\n",
      "Realized estimand type: nonparametric-ate\n",
      "Estimand expression:\n",
      "                                                                              \n",
      "Expectation(Derivative(income, voucher))⋅Expectation(Derivative(education, vou\n",
      "\n",
      "      -1\n",
      "cher))  \n",
      "Estimand assumption 1, As-if-random: If U→→income then ¬(U →→{voucher})\n",
      "Estimand assumption 2, Exclusion: If we remove {voucher}→{education}, then ¬({voucher}→income)\n",
      "Estimand assumption 3, treatment_effect_homogeneity: Each unit's treatment ['education'] is affected in the same way by common causes of ['education'] and income\n",
      "Estimand assumption 4, outcome_effect_homogeneity: Each unit's outcome income is affected in the same way by common causes of ['education'] and income\n",
      "\n",
      "Target units: ate\n",
      "\n",
      "## Estimate\n",
      "Mean value: 3.9971317330888536\n",
      "p-value: [0, 0.001]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Step 3: Estimate\n",
    "#Choose the second estimand: using IV\n",
    "estimate = model.estimate_effect(identified_estimand,\n",
    "        method_name=\"iv.instrumental_variable\", test_significance=True)\n",
    "\n",
    "print(estimate)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have an estimate, indicating that increasing `education` by one unit increases `income` by 4 points.  \n",
    "\n",
    "Next we check the robustness of the estimate using a Placebo refutation test. In this test, the treatment is replaced by an independent random variable (while preserving the correlation with the instrument), so that the true causal effect should be zero. We check if our estimator also provides the correct answer of zero. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Refute: Use a Placebo Treatment\n",
      "Estimated effect:3.9971317330888536\n",
      "New effect:0.0130069419635361\n",
      "p value:0.49\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Step 4: Refute\n",
    "ref = model.refute_estimate(identified_estimand, estimate, method_name=\"placebo_treatment_refuter\", placebo_type=\"permute\") # only permute placebo_type works with IV estimate\n",
    "print(ref)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The refutation gives confidence that the estimate is not capturing any noise in the data.\n",
    "\n",
    "Since this is simulated data, we also know the true causal effect is `4` (see the `income_education` parameter of the data-generating process above)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we show the same estimation by another method to verify the result from DoWhy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>IV2SLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>income</td>      <th>  R-squared:         </th> <td>   0.889</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                 <td>IV2SLS</td>      <th>  Adj. R-squared:    </th> <td>   0.889</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>               <td>Two Stage</td>    <th>  F-statistic:       </th> <td>   1368.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                    <td>Least Squares</td>  <th>  Prob (F-statistic):</th> <td>2.65e-189</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sun, 09 Jan 2022</td> <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:49:54</td>     <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  1000</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   998</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th> <td>   10.1524</td> <td>    1.009</td> <td>   10.059</td> <td> 0.000</td> <td>    8.172</td> <td>   12.133</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>education</th> <td>    3.9971</td> <td>    0.108</td> <td>   36.991</td> <td> 0.000</td> <td>    3.785</td> <td>    4.209</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 4.718</td> <th>  Durbin-Watson:     </th> <td>   2.068</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.095</td> <th>  Jarque-Bera (JB):  </th> <td>   4.731</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.168</td> <th>  Prob(JB):          </th> <td>  0.0939</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.979</td> <th>  Cond. No.          </th> <td>    26.4</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                          IV2SLS Regression Results                           \n",
       "==============================================================================\n",
       "Dep. Variable:                 income   R-squared:                       0.889\n",
       "Model:                         IV2SLS   Adj. R-squared:                  0.889\n",
       "Method:                     Two Stage   F-statistic:                     1368.\n",
       "                        Least Squares   Prob (F-statistic):          2.65e-189\n",
       "Date:                Sun, 09 Jan 2022                                         \n",
       "Time:                        11:49:54                                         \n",
       "No. Observations:                1000                                         \n",
       "Df Residuals:                     998                                         \n",
       "Df Model:                           1                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Intercept     10.1524      1.009     10.059      0.000       8.172      12.133\n",
       "education      3.9971      0.108     36.991      0.000       3.785       4.209\n",
       "==============================================================================\n",
       "Omnibus:                        4.718   Durbin-Watson:                   2.068\n",
       "Prob(Omnibus):                  0.095   Jarque-Bera (JB):                4.731\n",
       "Skew:                          -0.168   Prob(JB):                       0.0939\n",
       "Kurtosis:                       2.979   Cond. No.                         26.4\n",
       "==============================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "income_vec, endog = ps.dmatrices(\"income ~ education\", data=df)\n",
    "exog = ps.dmatrix(\"voucher\", data=df)\n",
    "\n",
    "m = IV2SLS(income_vec, endog, exog).fit()\n",
    "m.summary()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}