Smooth Estimators of Optimal Dynamic Treatment Regimes

• Activity Number: 456
• Type: Invited
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Kristin A. Linn*+ and Eric Laber and Len Stefanski
• Companies: NC State University and NC State University and North Carolina State University
• Keywords: Dynamic treatment regimes ; Personalized medicine ; Q-learning

Abstract

Clinicians wanting to form evidence-based rules for optimal treatment allocation over time have begun to estimate such rules using data collected from observational or randomized studies. Popular methods for estimating optimal sequential decision rules from data, such as Q-learning, are approximate dynamic programming algorithms that require modeling nonsmooth, nonmonotone transformations of the data. Unfortunately, postulating an adequately expressive, yet parsimonious, model for the transformed data is difficult. Under many simple generative models, the most commonly employed working models---namely, linear models---are misspecified. Furthermore, such estimators are nonregular, making statistical inference difficult. We propose an alternative strategy for estimating optimal sequential decision rules wherein all modeling takes place before nonsmooth, nonmonotone transformations of the data are applied. This simple interchange of modeling and transforming the data leads to high-quality estimated sequential decision rules and, in many cases, allows for simplified exploratory data analysis, model building and validation, and normal limit theory.