Robert W. Makuch Distinguished Lecture in Biostatistics
Susan A. Murphy
Mallinckrodt Professor of Statistics and Computer Science
Radcliffe Alumnae Professor at the Radcliffe Institute
Inference Using Adaptively Collected Data
Bandit algorithms are increasingly used in real-world sequential decision-making problems. Associated with this is an increased desire to be able to use the resulting datasets to answer scientific questions like: Did one type of ad lead to more purchases? In which contexts is a mobile health intervention effective? However, there is a lack of general methods for conducting statistical inference using more complex models on data collected with (contextual) bandit algorithms; for example, current methods cannot be used for valid inference on parameters in a logistic regression model for a binary reward. We discuss our work on theory justifying the use of M-estimators—which includes estimators based on empirical risk minimization as well as maximum likelihood—on data collected with adaptive algorithms, including(contextual) bandit algorithms. Specifically, we show that M-estimators, modified with particular adaptive weights, can be used to construct asymptotically valid confidence regions for a variety of inferential targets.
DATE: Wednesday, November 10, 2021
TIME: 4:00 p.m. EST
Susan A. Murphy is a Radcliffe Alumnae Professor at Harvard Radcliffe Institute and a professor of statistics and computer science at the Harvard John A. Paulson School of Engineering and Applied Sciences. A 2013 recipient of a MacArthur Fellowship, she was previously the H. E. Robbins Distinguished University Professor of Statistics, a research professor at the Institute for Social Research, and a professor of psychiatry, all at the University of Michigan.
Murphy earned her BS from Louisiana State University and her PhD from the University of North Carolina at Chapel Hill. Her research focuses on analytic methods to design and evaluate medical treatments that adapt to individuals, including some that use mobile devices to deliver tailored interventions for drug addicts, smokers, and heart disease patients, among others. She is a member of the National Academy of Medicine and of the National Academy of Sciences.
Susan A. Murphy awarded Van Wijngaarden Award (Harvard Gazette, 9/24/21)
Robert Makuch is a Professor in the Department of Biostatistics at the Yale School of Public Health and Director of the Regulatory Affairs Track. A graduate of the University of Connecticut (BA), University of Washington (MA – mathematics), and Yale University (MPhil, PhD), Professor Makuch worked at the National Cancer Institute (NCI) and the World Health Organization’s International Agency for Research on Cancer early in his career. He also worked for six months at the National Cancer Research Center in Tokyo, Japan.
He also was heavily involved in HIV research from the mid 80’s through the early-mid 90’s. He participated on the data monitoring committee for the original AZT vs. placebo randomized clinical trial in AIDS patients, and served on numerous committees for the NCI and the National Institute of Allergy and Infectious Diseases. He also worked closely with the Food and Drug Administration (FDA), developing and implementing more than 200 HIV studies. He also served as a Special Government Employee (SGE) to the FDA. He returned to Yale in 1986, and has worked extensively on methodologic issues in clinical trials and large population-based studies since. Another area of current interest involves detection of rare adverse drug events, especially in the post-marketing environment.
These areas of methodologic research evolved as a result of his continued interest (since the mid 1980s) in regulatory affairs science. In addition, Makuch developed a regulatory affairs track at YSPH for graduate and post-doctoral level students, and over the past 10 years has been the leader of more than 25 training programs for senior delegations of the Chinese Food and Drug Administration. His areas of medical application include cancer, HIV, arthritis, and cardiovascular disease.
In 2003, Makuch received the American Statistical Association Fellow Award for his numerous contributions to the field. In 2008, Makuch was received a Distinguished Alumni Award from the University of Connecticut. In 2012, Makuch was nominated to serve on the University of Connecticut Dean’s Advisory Board for the College of Liberal Arts and Sciences. He also has been a decades-long member of Phi Beta Kappa. He also developed a 5-year biostatistics training program in Japan, in collaboration with the Japanese government. His primary research interests continue to be methodologic issues in the design, conduct, analysis, and interpretation of clinical and large-population/epidemiologic studies. Design and sample size considerations for Phase IV studies is another active research area, in which a new class of hybrid designs has been proposed for scientific and regulatory purposes to detect rare adverse events.