{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Python: Policy Learning with Trees\n", "\n", "In this simple example, we illustrate how the [DoubleML](https://docs.doubleml.org/stable/index.html) package can be used to estimate deterministic binary treatment policies." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Data\n", "\n", "The data will be generated with a simple data generating process to enable us to know the true policy cut-offs." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import numpy as np \n", "import pandas as pd\n", "import doubleml as dml" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here, we consider a treatment effect that depends on two covariates which might correspond to opposite effects depending on age or income.\n", "For simplicity, the treatment effect within each group is generated to be constant." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "def group_effect(x):\n", " if x[0] <= -0.3:\n", " te = 2.2\n", " elif (x[0] >= 0.2) & (x[1] >= 0.4):\n", " te = 2\n", " else:\n", " te = -2\n", " return te" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The data is generated as\n", "\n", "\n", "\\begin{aligned}\n", "Y_i & = g(W_1,W_2)T_i + \\langle W_i,\\gamma_0\\rangle + \\epsilon_i \\\\\n", "T_i & = \\langle W_i,\\beta_0\\rangle +\\eta_i,\n", "\\end{aligned}\n", "\n", "\n", "where $W_i\\sim\\mathcal{N}(0,I_{d_w})$ and $\\epsilon_i,\\eta_i\\sim\\mathcal{U}[0,1]$.\n", "The coefficient vectors $\\gamma_0$ and $\\beta_0$ both have small random support which values are drawn independently from $\\mathcal{U}[0,1]$.\n", "Further, $g(w_1,w_2)$ defines the conditional treatment effect, which is generated depending on $\\{w_1,w_2\\}$.\n", "\n", "$$g(w_1) = \\begin{cases}2.2\\quad &\\text{for } w_1\\le -0.3\\\\\n", "2\\quad &\\text{for } w_1\\ge 0.2 \\land w_2\\ge 0.4\\\\\n", "-2\\quad &\\text{otherwise. } \\\\\n", " \\end{cases}$$" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "def create_synthetic_group_data(n_samples=200, n_w=10, support_size=5):\n", " \"\"\"\n", " Creates a simple synthetic example for group effects.\n", "\n", " Parameters\n", " ----------\n", " n_samples : int\n", " Number of samples.\n", " Default is 200.\n", "\n", " n_w : int\n", " Dimension of covariates.\n", " Default is 10.\n", "\n", " support_size : int\n", " Number of relevant covariates.\n", " Default is 5.\n", "\n", " Returns\n", " -------\n", " data : pd.DataFrame\n", " A data frame.\n", "\n", " \"\"\"\n", " # Outcome support\n", " support_w = np.random.choice(np.arange(n_w), size=support_size, replace=False)\n", " coefs_w = np.random.uniform(0, 1, size=support_size)\n", " # Define the function to generate the noise\n", " epsilon_sample = lambda n: np.random.uniform(-1, 1, size=n_samples)\n", " # Treatment support\n", " # Assuming the matrices gamma and beta have the same number of non-zero components\n", " support_t = np.random.choice(np.arange(n_w), size=support_size, replace=False)\n", " coefs_t = np.random.uniform(0, 1, size=support_size)\n", " # Define the function to generate the noise\n", " eta_sample = lambda n: np.random.uniform(-1, 1, size=n_samples)\n", "\n", " # Generate controls, covariates, treatments and outcomes\n", " w = np.random.normal(0, 1, size=(n_samples, n_w))\n", " # Group treatment effect\n", " te = np.apply_along_axis(group_effect, axis=1, arr=w)\n", " # Define treatment\n", " log_odds = np.dot(w[:, support_t], coefs_t) + eta_sample(n_samples)\n", " t_sigmoid = 1 / (1 + np.exp(-log_odds))\n", " t = np.array([np.random.binomial(1, p) for p in t_sigmoid])\n", " # Define the outcome\n", " y = te * t + np.dot(w[:, support_w], coefs_w) + epsilon_sample(n_samples)\n", "\n", " # Now we build the dataset\n", " y_df = pd.DataFrame({'y': y})\n", " t_df = pd.DataFrame({'t': t})\n", " w_df = pd.DataFrame(data=w, index=np.arange(w.shape[0]), \n", " columns=[f'w_{i}' for i in range(1, w.shape[1] + 1)])\n", "\n", " data = pd.concat([y_df, t_df, w_df], axis=1)\n", " covariates = list(w_df.columns.values)\n", "\n", " return data, covariates" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We will consider a quite small number of covariates to ensure fast calcualtion." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "# DGP constants\n", "np.random.seed(42)\n", "n_samples = 500\n", "n_w = 10\n", "support_size = 5\n", "\n", "# Create data\n", "data, covariates = create_synthetic_group_data(n_samples=n_samples, n_w=n_w, support_size=support_size)\n", "data_dml_base = dml.DoubleMLData(data,\n", " y_col='y',\n", " d_cols='t',\n", " x_cols=covariates)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Interactive Regression Model (IRM)\n", "The first step is to fit a [DoubleML IRM Model](https://docs.doubleml.org/stable/guide/models.html#interactive-regression-model-irm) to the data." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training IRM Model\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# First stage estimation\n", "from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor\n", "randomForest_reg = RandomForestRegressor(n_estimators=200, random_state=42)\n", "randomForest_class = RandomForestClassifier(n_estimators=200, random_state=42)\n", "\n", "np.random.seed(42)\n", "\n", "dml_irm = dml.DoubleMLIRM(data_dml_base,\n", " ml_g=randomForest_reg,\n", " ml_m=randomForest_class,\n", " trimming_threshold=0.01,\n", " n_folds=5)\n", "print(\"Training IRM Model\")\n", "dml_irm.fit(store_predictions=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Policy Learning with Trees\n", "Next, we specify the covariates as a [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html) against which to learn the policy. This can either be all covariates $w_i$, or if domain knowledge is available we can use a subset." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " w_1 w_2\n", "0 -0.428046 -0.742407\n", "1 -0.600254 0.947440\n", "2 -1.265119 1.091992\n", "3 -1.346678 -0.880591\n", "4 -1.508153 1.099647\n", ".. ... ...\n", "495 -2.072293 -0.951920\n", "496 0.144908 0.280963\n", "497 -0.877455 -2.404550\n", "498 -0.981104 -0.830301\n", "499 -0.555445 0.021866\n", "\n", "[500 rows x 2 columns]\n" ] } ], "source": [ "features = data[[\"w_1\",\"w_2\"]].copy()\n", "print(features)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To estimate a Policy just call the policy_tree() method and supply the DataFrame with the features and the depth of the desired tree." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "