Optimal Policy Learning for Multi-Action Treatment with Risk Preference using Stata

2025/09/08 by Giovanni Cerulli, Cerulli, Giovanni
Economics, Econometrics and Finance · #Econometrics (econ.EM) #FOS: Economics and business #Healthcare Policy and Management

paper · pdf · doi:10.48550/arxiv.2509.06851

openalex publication_date 2025/09/08 · openalex created_date 2025/10/11 · openalex updated_date 2026/07/28

Abstract

This paper presents the Stata community-distributed command "opl_ma_fb" (and the companion command "opl_ma_vf"), for implementing the first-best Optimal Policy Learning (OPL) algorithm to estimate the best treatment assignment given the observation of an outcome, a multi-action (or multi-arm) treatment, and a set of observed covariates (features). It allows for different risk preferences in decision-making (i.e., risk-neutral, linear risk-averse, and quadratic risk-averse), and provides a graphical representation of the optimal policy, along with an estimate of the maximal welfare (i.e., the value-function estimated at optimal policy) using regression adjustment (RA), inverse-probability weighting (IPW), and doubly robust (DR) formulas.

Citations

• Optimal Policy Learning with Observational Data in Multi-Action Scenarios: Estimation, Risk Preference, and Potential Failures
• High-Confidence Off-Policy (or Counterfactual) Variance Estimation
• Self-Supervised Reinforcement Learning for Recommender Systems
• Policy Learning With Observational Data
• From Predictive to Prescriptive Analytics
• Doubly Robust Policy Evaluation and Learning

Related

• Optimal Policy Learning with Observational Data in Multi-Action Scenarios: Estimation, Risk Preference, and Potential Failures
• Policy Learning with Distributional Welfare
• Risk-adjusted policy learning and the social cost of uncertainty: Theory and evidence from CAP evaluation
• Policy Learning for Optimal Dynamic Treatment Regimes with Observational Data
• Optimal Policy Learning: From Theory to Practice