policytree: Policy learning via doubly robust empirical welfare maximization over trees

A package for learning optimal policies via doubly robust empirical welfare maximization over trees. This package implements the multi-action doubly robust approach of Zhou, Athey and Wager (2018) in the case where we want to learn policies that belong to the class of depth k decision trees. Many practical policy applications require interpretable predictions. For example, a drug prescription guide that follows a simple 2-question Yes/No checklist can be encoded as a depth 2 decision tree (does the patient have a heart condition - etc.). policytree currently has support for estimating multi-action treatment effects with one vs. all grf, calculating statistics such as double robust scores (support for a subset of grf forest types) and fitting optimal policies with exact tree search.