AI & Machine Learning

BalanceHD is a package for estimation of average treatment effects in high dimensions via approximate residual balancing.

CausalTree is a package for building binary regression tree models in two stages focusing on estimating heterogeneous causal treatment effects.

DS-WGAN is a Python package built on PyTorch for using WGANs to simulate from joint and conditional distributions of economic datasets.

Generalized Random Forests Package (GRF Package) is an R package for forest-based statistical estimation and inference. It provides methods for causal effect estimation and a framework to create forests for new statistical tasks, with new features and methods regularly added.

The Lab's Generalized Random Forests (GRF) work has been adopted by Microsoft’s “EconML” toolkit with machine learning resources for economists.

MCPanel is a package that provides functions to fit a low-rank model to a partially observed matrix.

Multi-Armed Qini is a package for policy evaluation using generalized Qini curves: Evaluate data-driven treatment targeting rules for one or more treatment arms over different budget constraints in experimental or observational settings under unconfoundedness.

ParTreat is software for estimating treatment effects in experiments where the outcome distributions may have “fat tails,” i.e., where understanding extremely high or low outcomes is important for policy but analyzing them makes the analysis of experiments very noisy.

PolicyTree is a package that implements the approach of learning optimal policies through doubly robust estimation of policy trees.

Package for implementing methods for providing sufficient representations of categorical variables

Synthdid is a package that implements the synthetic difference in difference estimator for the average treatment effect in panel data.

Torch Choice is a library for flexible, fast choice modeling with PyTorch designed for both estimation and prediction.

This report is designed for any audience looking to learn more about how machine learning can add new techniques to behavioral design, causal inference, and experimentation.

This tutorial introduces key concepts in machine learning-based causal inference and is an ongoing project with new chapters uploaded as they are completed. Topics currently covered:

• Introduction to Machine Learning
• Average Treatment Effects
• Heterogeneous Treatment Effects
• Policy Evaluation
• Policy Learning

This course is a series of videos designed for any audience looking to learn more about how machine learning can be used to measure the effects of interventions, understand the heterogeneous impact of interventions, and design targeted treatment assignment policies.