American Statistical Association
A dynamic treatment regimen effectively incorporates both accrued information and long-term effects of treatment from specially designed clinical trials. As these become more and more popular in conjunction with longitudinal data from clinical studies, the development of statistical inference for optimal dynamic treatment regimens is a high priority. This is very challenging due to the difficulties arising from non-regularities in the treatment effect parameters. In this talk, we propose a new machine learning framework called penalized Q-learning (PQ-learning), under which the non-regularities can be resolved and valid statistical inference established. We also propose a new statistical procedure -- individual selection -- and corresponding methods for incorporating individual selection within PQ-learning. Extensive numerical studies are presented which compare the proposed methods with existing methods, under a variety of non-regular scenarios, and demonstrate that the proposed approach is both inferentially and computationally superior. The proposed method is demonstrated with the data from a depression clinical trial study.
Rui Song got her Ph.D. from University of Wisconsin in 2006. She is an assistant professor of Statistics at Colorado State University. Her current research interests include high-dimensional statistical learning, dynamic treatment regimens and personalized medicine, semiparametric inference and empirical processes.
|Date:||Thursday, February 23, 2012|
|Time:||4:00 - 5:00 P.M.|
Mailman School of Public Health
Department of Biostatistics
722 West 168th Street
New York, New York