NIAAA awarded $0.42 million grant to Dr. Ertefaie to develop individualized treatment strategies for controlling alcohol use. Our overarching aim is to address the need for robust, rigorous and efficient methods for estimating optimal treatment strategies in high-dimensional settings. Current methods for constructing individualized treatment strategies rely on certain modeling assumptions, and thus, the results can be very sensitive to the postulated models. The R21 aims to relax these unrealistic assumptions by leveraging the state-of-the-art nonparametric regression methods. We also investigate a novel technique to identify the important set of treatment effect modifiers among a long list of candidate variables. This work will also help to pave the way for future studies that advance personalized medicine, enabling the discovery of genetic and phenotypic subgroups that respond favorably to experimental treatments.
Robust Q-learning
Ertefaie A, McKay J. R., Oslin D., and Strawderman R. L. (2020). Journal of the American Statistical Association, DOI: 10.1080/01621459.2020.1753522 Q-learning is a regression-based approach that is