{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e1d1e898",
   "metadata": {},
   "source": [
    "# Intro to DoWhy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5bef9c3d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Path to dataset files: /Users/hunter.lybbert/.cache/kagglehub/datasets/tbyrnes/advertising/versions/1\n"
     ]
    }
   ],
   "source": [
    "import kagglehub\n",
    "\n",
    "# Download latest version\n",
    "path = kagglehub.dataset_download(\"tbyrnes/advertising\")\n",
    "\n",
    "print(\"Path to dataset files:\", path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f81751e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "24708f57",
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Ad Topic Line</th>\n",
       "      <th>City</th>\n",
       "      <th>Male</th>\n",
       "      <th>Country</th>\n",
       "      <th>Timestamp</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>68.95</td>\n",
       "      <td>35</td>\n",
       "      <td>61833.90</td>\n",
       "      <td>256.09</td>\n",
       "      <td>Cloned 5thgeneration orchestration</td>\n",
       "      <td>Wrightburgh</td>\n",
       "      <td>0</td>\n",
       "      <td>Tunisia</td>\n",
       "      <td>2016-03-27 00:53:11</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>80.23</td>\n",
       "      <td>31</td>\n",
       "      <td>68441.85</td>\n",
       "      <td>193.77</td>\n",
       "      <td>Monitored national standardization</td>\n",
       "      <td>West Jodi</td>\n",
       "      <td>1</td>\n",
       "      <td>Nauru</td>\n",
       "      <td>2016-04-04 01:39:02</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>69.47</td>\n",
       "      <td>26</td>\n",
       "      <td>59785.94</td>\n",
       "      <td>236.50</td>\n",
       "      <td>Organic bottom-line service-desk</td>\n",
       "      <td>Davidton</td>\n",
       "      <td>0</td>\n",
       "      <td>San Marino</td>\n",
       "      <td>2016-03-13 20:35:42</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>74.15</td>\n",
       "      <td>29</td>\n",
       "      <td>54806.18</td>\n",
       "      <td>245.89</td>\n",
       "      <td>Triple-buffered reciprocal time-frame</td>\n",
       "      <td>West Terrifurt</td>\n",
       "      <td>1</td>\n",
       "      <td>Italy</td>\n",
       "      <td>2016-01-10 02:31:19</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68.37</td>\n",
       "      <td>35</td>\n",
       "      <td>73889.99</td>\n",
       "      <td>225.58</td>\n",
       "      <td>Robust logistical utilization</td>\n",
       "      <td>South Manuel</td>\n",
       "      <td>0</td>\n",
       "      <td>Iceland</td>\n",
       "      <td>2016-06-03 03:36:18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Daily Time Spent on Site  Age  Area Income  Daily Internet Usage  \\\n",
       "0                     68.95   35     61833.90                256.09   \n",
       "1                     80.23   31     68441.85                193.77   \n",
       "2                     69.47   26     59785.94                236.50   \n",
       "3                     74.15   29     54806.18                245.89   \n",
       "4                     68.37   35     73889.99                225.58   \n",
       "\n",
       "                           Ad Topic Line            City  Male     Country  \\\n",
       "0     Cloned 5thgeneration orchestration     Wrightburgh     0     Tunisia   \n",
       "1     Monitored national standardization       West Jodi     1       Nauru   \n",
       "2       Organic bottom-line service-desk        Davidton     0  San Marino   \n",
       "3  Triple-buffered reciprocal time-frame  West Terrifurt     1       Italy   \n",
       "4          Robust logistical utilization    South Manuel     0     Iceland   \n",
       "\n",
       "             Timestamp  Clicked on Ad  \n",
       "0  2016-03-27 00:53:11              0  \n",
       "1  2016-04-04 01:39:02              0  \n",
       "2  2016-03-13 20:35:42              0  \n",
       "3  2016-01-10 02:31:19              0  \n",
       "4  2016-06-03 03:36:18              0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_df = pd.read_csv(\"/Users/hunter.lybbert/.cache/kagglehub/datasets/tbyrnes/advertising/versions/1/advertising.csv\")\n",
    "raw_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "024c792e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['daily_time_spent_on_site', 'age', 'area_income',\n",
       "       'daily_internet_usage', 'ad_topic_line', 'city', 'male', 'country',\n",
       "       'timestamp', 'clicked_on_ad'],\n",
       "      dtype='str')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_df.rename(columns=lambda x:x.replace(' ', '_').lower(),inplace=True)\n",
    "raw_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ad83eb50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "daily_time_spent_on_site    float64\n",
       "age                           int64\n",
       "area_income                 float64\n",
       "daily_internet_usage        float64\n",
       "ad_topic_line                   str\n",
       "city                            str\n",
       "male                          int64\n",
       "country                         str\n",
       "clicked_on_ad                 int64\n",
       "day_of_week                   int32\n",
       "hour                          int32\n",
       "month                         int32\n",
       "dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = raw_df.copy(deep=True)\n",
    "df.timestamp = pd.to_datetime(df.timestamp)\n",
    "df[\"day_of_week\"] = df[\"timestamp\"].dt.dayofweek\n",
    "df[\"hour\"] = df[\"timestamp\"].dt.hour\n",
    "df[\"month\"] = df[\"timestamp\"].dt.month\n",
    "df.drop(columns=[\"timestamp\"], inplace=True)\n",
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a7445808",
   "metadata": {},
   "outputs": [],
   "source": [
    "# age -> daily_time_spent_on_site;\n",
    "# age -> daily_internet_usage;\n",
    "# male -> daily_internet_usage;\n",
    "# male -> daily_time_spent_on_site;\n",
    "# daily_time_spent_on_site -> timestamp;\n",
    "# daily_internet_usage -> timestamp;\n",
    "# daily_time_spent_on_site -> clicked_on_ad;\n",
    "# daily_internet_usage -> clicked_on_ad;\n",
    "# country -> area_income -> clicked_on_ad;\n",
    "# country -> daily_internet_usage;\n",
    "# area_income -> daily_time_spent_on_site;\n",
    "# area_income -> daily_internet_usage;\n",
    "# area_income -> clicked_on_ad;\n",
    "# city -> country -> area_income -> clicked_on_ad;\n",
    "# city -> area_income;\n",
    "\n",
    "# 'daily_internet_usage',\n",
    "# 'ad_topic_line',\n",
    "# 'city',\n",
    "# 'male',\n",
    "# 'country',\n",
    "# 'timestamp',\n",
    "# 'clicked_on_ad'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "0eff76b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "causal_graph = \"\"\"\n",
    "digraph {\n",
    "daily_time_spent_on_site;\n",
    "age;\n",
    "area_income;\n",
    "daily_internet_usage;\n",
    "ad_topic_line;\n",
    "city;\n",
    "male;\n",
    "country;\n",
    "day_of_week;\n",
    "hour;\n",
    "month;\n",
    "clicked_on_ad;\n",
    "age -> daily_time_spent_on_site;\n",
    "age -> daily_internet_usage;\n",
    "male -> daily_internet_usage;\n",
    "male -> daily_time_spent_on_site;\n",
    "daily_time_spent_on_site -> day_of_week;\n",
    "daily_internet_usage -> day_of_week;\n",
    "daily_time_spent_on_site -> month;\n",
    "daily_internet_usage -> month;\n",
    "daily_time_spent_on_site -> hour;\n",
    "daily_internet_usage -> hour;\n",
    "country -> area_income -> clicked_on_ad;\n",
    "country -> daily_internet_usage;\n",
    "area_income -> daily_time_spent_on_site;\n",
    "area_income -> daily_internet_usage;\n",
    "area_income -> clicked_on_ad;\n",
    "city -> country -> area_income -> clicked_on_ad;\n",
    "city -> area_income;\n",
    "day_of_week -> clicked_on_ad;\n",
    "hour -> clicked_on_ad;\n",
    "month -> clicked_on_ad;\n",
    "}\n",
    "\"\"\"\n",
    "# ad_topic_line -> clicked_on_ad;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3ef90c2b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from dowhy import CausalModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "10a8e9c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = CausalModel(\n",
    "    data=df,\n",
    "    graph=causal_graph.replace(\"\\n\", \" \"),\n",
    "    treatment=\"hour\",\n",
    "    outcome=\"clicked_on_ad\",\n",
    ")\n",
    "model.view_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4700ef39",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                Cloned 5thgeneration orchestration\n",
       "1                Monitored national standardization\n",
       "2                  Organic bottom-line service-desk\n",
       "3             Triple-buffered reciprocal time-frame\n",
       "4                     Robust logistical utilization\n",
       "5                   Sharable client-driven software\n",
       "6                        Enhanced dedicated support\n",
       "7                          Reactive local challenge\n",
       "8                    Configurable coherent function\n",
       "9                Mandatory homogeneous architecture\n",
       "10                   Centralized neutral neural-net\n",
       "11    Team-oriented grid-enabled Local Area Network\n",
       "12            Centralized content-based focus group\n",
       "13                 Synergistic fresh-thinking array\n",
       "14                    Grass-roots coherent extranet\n",
       "15               Persistent demand-driven interface\n",
       "16               Customizable multi-tasking website\n",
       "17                       Intuitive dynamic attitude\n",
       "18     Grass-roots solution-oriented conglomeration\n",
       "19                       Advanced 24/7 productivity\n",
       "20            Object-based reciprocal knowledgebase\n",
       "21                Streamlined non-volatile analyzer\n",
       "22            Mandatory disintermediate utilization\n",
       "23               Future-proofed methodical protocol\n",
       "24                    Exclusive neutral parallelism\n",
       "Name: ad_topic_line, dtype: str"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# df.ad_topic_line.head(25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f9f5e508",
   "metadata": {},
   "outputs": [],
   "source": [
    "# model = CausalModel(\n",
    "#     data=df,\n",
    "#     graph=causal_graph.replace(\"\\n\", \" \"),\n",
    "#     treatment=\"timestamp\",\n",
    "#     outcome=\"clicked_on_ad\",\n",
    "# )\n",
    "# model.view_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "88da5645",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "   d                                                                   \n",
      "───────(E[clicked_on_ad|daily_internet_usage,daily_time_spent_on_site])\n",
      "d[hour]                                                                \n",
      "Estimand assumption 1, Unconfoundedness: If U→{hour} and U→clicked_on_ad then P(clicked_on_ad|hour,daily_internet_usage,daily_time_spent_on_site,U) = P(clicked_on_ad|hour,daily_internet_usage,daily_time_spent_on_site)\n",
      "\n",
      "### Estimand : 2\n",
      "Estimand name: iv\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 3\n",
      "Estimand name: frontdoor\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 4\n",
      "Estimand name: general_adjustment\n",
      "Estimand expression:\n",
      "   d                                                                   \n",
      "───────(E[clicked_on_ad|daily_internet_usage,daily_time_spent_on_site])\n",
      "d[hour]                                                                \n",
      "Estimand assumption 1, Unconfoundedness: If U→{hour} and U→clicked_on_ad then P(clicked_on_ad|hour,daily_internet_usage,daily_time_spent_on_site,U) = P(clicked_on_ad|hour,daily_internet_usage,daily_time_spent_on_site)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "estimands = model.identify_effect()\n",
    "print(estimands)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "7b664bd2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimate of causal effect: *** Causal Estimate ***\n",
      "\n",
      "## Identified estimand\n",
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "   d                                                                   \n",
      "───────(E[clicked_on_ad|daily_internet_usage,daily_time_spent_on_site])\n",
      "d[hour]                                                                \n",
      "Estimand assumption 1, Unconfoundedness: If U→{hour} and U→clicked_on_ad then P(clicked_on_ad|hour,daily_internet_usage,daily_time_spent_on_site,U) = P(clicked_on_ad|hour,daily_internet_usage,daily_time_spent_on_site)\n",
      "\n",
      "## Realized estimand\n",
      "b: clicked_on_ad~hour+daily_internet_usage+daily_time_spent_on_site+hour*day_of_week+hour*month\n",
      "Target units: \n",
      "\n",
      "## Estimate\n",
      "Mean value: -0.00046934978493284873\n",
      "p-value: [0.74907975]\n",
      "95.0% confidence interval: (np.float64(-0.002546200086192607), np.float64(0.0018146626893807971))\n",
      "### Conditional Estimates\n",
      "__categorical__day_of_week  __categorical__month\n",
      "(-0.001, 1.0]               (0.999, 2.0]           -0.000541\n",
      "                            (2.0, 3.0]             -0.000529\n",
      "                            (3.0, 4.0]             -0.000526\n",
      "                            (4.0, 6.0]             -0.000513\n",
      "                            (6.0, 7.0]             -0.000502\n",
      "(1.0, 2.0]                  (0.999, 2.0]           -0.000507\n",
      "                            (2.0, 3.0]             -0.000498\n",
      "                            (3.0, 4.0]             -0.000491\n",
      "                            (4.0, 6.0]             -0.000480\n",
      "                            (6.0, 7.0]             -0.000470\n",
      "(2.0, 4.0]                  (0.999, 2.0]           -0.000475\n",
      "                            (2.0, 3.0]             -0.000467\n",
      "                            (3.0, 4.0]             -0.000457\n",
      "                            (4.0, 6.0]             -0.000448\n",
      "                            (6.0, 7.0]             -0.000438\n",
      "(4.0, 5.0]                  (0.999, 2.0]           -0.000445\n",
      "                            (2.0, 3.0]             -0.000434\n",
      "                            (3.0, 4.0]             -0.000427\n",
      "                            (4.0, 6.0]             -0.000415\n",
      "                            (6.0, 7.0]             -0.000406\n",
      "(5.0, 6.0]                  (0.999, 2.0]           -0.000422\n",
      "                            (2.0, 3.0]             -0.000412\n",
      "                            (3.0, 4.0]             -0.000405\n",
      "                            (4.0, 6.0]             -0.000395\n",
      "                            (6.0, 7.0]             -0.000384\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "estimate = model.estimate_effect(\n",
    "    identified_estimand=estimands,\n",
    "    method_name=\"backdoor.linear_regression\",\n",
    "    # method_name=\"backdoor.econml.dml.CausalForestDML\",\n",
    "    confidence_intervals=True,\n",
    "    test_significance=True,\n",
    "    # method_params={\n",
    "    #     \"init_params\": {\n",
    "    #         \"n_estimators\": 100,\n",
    "    #         \"random_state\": 42\n",
    "    #     }\n",
    "    # }\n",
    ")\n",
    "\n",
    "print(f'Estimate of causal effect: {estimate}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "07555590",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "causal_graph_updated = \"\"\"\n",
    "digraph {\n",
    "daily_time_spent_on_site;\n",
    "age;\n",
    "area_income;\n",
    "daily_internet_usage;\n",
    "ad_topic_line;\n",
    "city;\n",
    "male;\n",
    "country;\n",
    "day_of_week;\n",
    "hour;\n",
    "month;\n",
    "clicked_on_ad;\n",
    "age -> daily_time_spent_on_site;\n",
    "age -> daily_internet_usage;\n",
    "male -> daily_internet_usage;\n",
    "male -> daily_time_spent_on_site;\n",
    "daily_time_spent_on_site -> clicked_on_ad;\n",
    "daily_internet_usage -> clicked_on_ad;\n",
    "country -> area_income -> clicked_on_ad;\n",
    "country -> daily_internet_usage;\n",
    "area_income -> daily_time_spent_on_site;\n",
    "area_income -> daily_internet_usage;\n",
    "area_income -> clicked_on_ad;\n",
    "city -> country -> area_income -> clicked_on_ad;\n",
    "city -> area_income;\n",
    "}\n",
    "\"\"\"\n",
    "# ad_topic_line -> clicked_on_ad;\n",
    "\n",
    "model_updated = CausalModel(\n",
    "    data=df,\n",
    "    graph=causal_graph_updated.replace(\"\\n\", \" \"),\n",
    "    treatment=\"daily_internet_usage\",\n",
    "    outcome=\"clicked_on_ad\",\n",
    ")\n",
    "model_updated.view_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "b2e14516",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "           d                                                                  \n",
      "───────────────────────(E[clicked_on_ad|area_income,daily_time_spent_on_site])\n",
      "d[daily_internet_usage]                                                       \n",
      "Estimand assumption 1, Unconfoundedness: If U→{daily_internet_usage} and U→clicked_on_ad then P(clicked_on_ad|daily_internet_usage,area_income,daily_time_spent_on_site,U) = P(clicked_on_ad|daily_internet_usage,area_income,daily_time_spent_on_site)\n",
      "\n",
      "### Estimand : 2\n",
      "Estimand name: iv\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 3\n",
      "Estimand name: frontdoor\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 4\n",
      "Estimand name: general_adjustment\n",
      "Estimand expression:\n",
      "           d                                                  \n",
      "───────────────────────(E[clicked_on_ad|age,area_income,male])\n",
      "d[daily_internet_usage]                                       \n",
      "Estimand assumption 1, Unconfoundedness: If U→{daily_internet_usage} and U→clicked_on_ad then P(clicked_on_ad|daily_internet_usage,age,area_income,male,U) = P(clicked_on_ad|daily_internet_usage,age,area_income,male)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "estimands_updated = model_updated.identify_effect()\n",
    "print(estimands_updated)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3165cda1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimate of causal effect: *** Causal Estimate ***\n",
      "\n",
      "## Identified estimand\n",
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "           d                                                                  \n",
      "───────────────────────(E[clicked_on_ad|area_income,daily_time_spent_on_site])\n",
      "d[daily_internet_usage]                                                       \n",
      "Estimand assumption 1, Unconfoundedness: If U→{daily_internet_usage} and U→clicked_on_ad then P(clicked_on_ad|daily_internet_usage,area_income,daily_time_spent_on_site,U) = P(clicked_on_ad|daily_internet_usage,area_income,daily_time_spent_on_site)\n",
      "\n",
      "## Realized estimand\n",
      "b: clicked_on_ad~daily_internet_usage+area_income+daily_time_spent_on_site+daily_internet_usage*daily_time_spent_on_site\n",
      "Target units: \n",
      "\n",
      "## Estimate\n",
      "Mean value: -0.005637221868382003\n",
      "p-value: [0.41300316]\n",
      "95.0% confidence interval: (np.float64(-0.006068843513229671), np.float64(-0.005152822316435213))\n",
      "### Conditional Estimates\n",
      "__categorical__daily_time_spent_on_site\n",
      "(32.599000000000004, 47.23]   -0.003223\n",
      "(47.23, 62.26]                -0.004660\n",
      "(62.26, 72.952]               -0.005953\n",
      "(72.952, 79.982]              -0.006809\n",
      "(79.982, 91.43]               -0.007554\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "estimate = model_updated.estimate_effect(\n",
    "    identified_estimand=estimands_updated,\n",
    "    method_name=\"backdoor.linear_regression\",\n",
    "    # method_name=\"backdoor.econml.dml.CausalForestDML\",\n",
    "    confidence_intervals=True,\n",
    "    test_significance=True,\n",
    "    # method_params={\n",
    "    #     \"init_params\": {\n",
    "    #         \"n_estimators\": 100,\n",
    "    #         \"random_state\": 42\n",
    "    #     }\n",
    "    # }\n",
    ")\n",
    "\n",
    "print(f'Estimate of causal effect: {estimate}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "1f087cba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "            d                                                                \n",
      "──────────────────────────(E[clicked_on_ad|area_income,daily_internet_usage])\n",
      "d[dailyₜᵢₘₑ ₛₚₑₙₜ ₒₙ ₛᵢₜₑ]                                                   \n",
      "Estimand assumption 1, Unconfoundedness: If U→{daily_time_spent_on_site} and U→clicked_on_ad then P(clicked_on_ad|daily_time_spent_on_site,area_income,daily_internet_usage,U) = P(clicked_on_ad|daily_time_spent_on_site,area_income,daily_internet_usage)\n",
      "\n",
      "### Estimand : 2\n",
      "Estimand name: iv\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 3\n",
      "Estimand name: frontdoor\n",
      "No such variable(s) found!\n",
      "\n",
      "### Estimand : 4\n",
      "Estimand name: general_adjustment\n",
      "Estimand expression:\n",
      "            d                                                    \n",
      "──────────────────────────(E[clicked_on_ad|age,area_income,male])\n",
      "d[dailyₜᵢₘₑ ₛₚₑₙₜ ₒₙ ₛᵢₜₑ]                                       \n",
      "Estimand assumption 1, Unconfoundedness: If U→{daily_time_spent_on_site} and U→clicked_on_ad then P(clicked_on_ad|daily_time_spent_on_site,age,area_income,male,U) = P(clicked_on_ad|daily_time_spent_on_site,age,area_income,male)\n",
      "\n",
      "Estimate of causal effect: *** Causal Estimate ***\n",
      "\n",
      "## Identified estimand\n",
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "            d                                                                \n",
      "──────────────────────────(E[clicked_on_ad|area_income,daily_internet_usage])\n",
      "d[dailyₜᵢₘₑ ₛₚₑₙₜ ₒₙ ₛᵢₜₑ]                                                   \n",
      "Estimand assumption 1, Unconfoundedness: If U→{daily_time_spent_on_site} and U→clicked_on_ad then P(clicked_on_ad|daily_time_spent_on_site,area_income,daily_internet_usage,U) = P(clicked_on_ad|daily_time_spent_on_site,area_income,daily_internet_usage)\n",
      "\n",
      "## Realized estimand\n",
      "b: clicked_on_ad~daily_time_spent_on_site+area_income+daily_internet_usage+daily_time_spent_on_site*daily_internet_usage\n",
      "Target units: \n",
      "\n",
      "## Estimate\n",
      "Mean value: -0.01407526806184456\n",
      "p-value: [0.16069722]\n",
      "95.0% confidence interval: (np.float64(-0.015233356611323856), np.float64(-0.012781125853657649))\n",
      "### Conditional Estimates\n",
      "__categorical__daily_internet_usage\n",
      "(104.779, 132.366]   -0.008043\n",
      "(132.366, 163.44]    -0.010821\n",
      "(163.44, 198.948]    -0.014298\n",
      "(198.948, 224.836]   -0.017275\n",
      "(224.836, 269.96]    -0.019939\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "model_updated_2 = CausalModel(\n",
    "    data=df,\n",
    "    graph=causal_graph_updated.replace(\"\\n\", \" \"),\n",
    "    treatment=\"daily_time_spent_on_site\",\n",
    "    outcome=\"clicked_on_ad\",\n",
    ")\n",
    "\n",
    "estimands_updated = model_updated_2.identify_effect()\n",
    "print(estimands_updated)\n",
    "\n",
    "estimate = model_updated_2.estimate_effect(\n",
    "    identified_estimand=estimands_updated,\n",
    "    method_name=\"backdoor.linear_regression\",\n",
    "    # method_name=\"backdoor.econml.dml.CausalForestDML\",\n",
    "    confidence_intervals=True,\n",
    "    test_significance=True,\n",
    "    # method_params={\n",
    "    #     \"init_params\": {\n",
    "    #         \"n_estimators\": 100,\n",
    "    #         \"random_state\": 42\n",
    "    #     }\n",
    "    # }\n",
    ")\n",
    "\n",
    "print(f'Estimate of causal effect: {estimate}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91e5e6a7",
   "metadata": {},
   "source": [
    "Start here by refuting the estimates that you developed, rework naming from the top so it is clear which models, graphs, and estimates/estimands are from which version of the modeling setup."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "faa37e6f",
   "metadata": {},
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (3983541303.py, line 1)",
     "output_type": "error",
     "traceback": [
      "  \u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[36]\u001b[39m\u001b[32m, line 1\u001b[39m\n\u001b[31m    \u001b[39m\u001b[31mFrom Claude:\u001b[39m\n         ^\n\u001b[31mSyntaxError\u001b[39m\u001b[31m:\u001b[39m invalid syntax\n"
     ]
    }
   ],
   "source": [
    "# From Claude:\n",
    "# 1. Placebo treatment refuter — randomizes the treatment; a real effect should vanish.\n",
    "#    Most diagnostic given your CI/p-value disagreement: if a *placebo* treatment also\n",
    "#    produces a \"significant\"-looking estimate under this same estimator, the significance\n",
    "#    test itself is unreliable, not just this particular result.\n",
    "res_placebo = model_updated.refute_estimate(\n",
    "    estimands_updated,\n",
    "    estimate,\n",
    "    method_name=\"placebo_treatment_refuter\",\n",
    "    placebo_type=\"permute\",\n",
    "    num_simulations=100,\n",
    ")\n",
    "print(res_placebo)\n",
    "\n",
    "# 2. Random common cause — adds an independent random covariate as a confounder.\n",
    "#    A robust estimate should barely move.\n",
    "res_random_cc = model_updated.refute_estimate(\n",
    "    estimands_updated,\n",
    "    estimate,\n",
    "    method_name=\"random_common_cause\",\n",
    "    num_simulations=100,\n",
    ")\n",
    "print(res_random_cc)\n",
    "\n",
    "# 3. Data subset refuter — refits on random subsets of the data.\n",
    "#    A stable effect should stay close to the full-sample estimate across subsets.\n",
    "res_subset = model_updated.refute_estimate(\n",
    "    estimands_updated,\n",
    "    estimate,\n",
    "    method_name=\"data_subset_refuter\",\n",
    "    subset_fraction=0.8,\n",
    "    num_simulations=100,\n",
    ")\n",
    "print(res_subset)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ef477c48",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "causal-inference",
   "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": "3.11.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}