James Robins, IQSS affiliate and faculty member in the Harvard School of Public Health, has been honored with an entire special issue of Statistical Science.
Robins, Mitchell L. and Robin LaFoley Dong Professor of Epidemiology and Professor of Biostatistics, is responsible for some of the most important foundational concepts and analytic methods in the fast-growing field of causal inference from observational and experimental data. The November 2014 Statistical Science, edited by Thomas S. Richardson and Andrea Rotnitzky, is devoted to papers on causal inference and semiparametric models, topics that have been the focus of Robins’ research and to which he has made notable contributions throughout his career. The papers are written by authors who have collaborated with or were otherwise influenced by Robins, including fellow IQSS faculty affiliate Tyler VanderWeele.
Robins is best known for developing methods for drawing causal inferences from complex observational and randomized studies with time-varying treatments or exposure. These methods, collectively referred to as g-methods, include inverse probability of treatment weighted and doubly robust estimation of marginal structural models, g-Estimation of Structural Nested Models, and the parametric g-formula. Over a career of over 30 years, Robins has applied his methods to research in such topics as AIDS treatment, smoking, post-menopausal estrogens, and arsenic exposure among metal workers.
A full issue inspired by Robins’ research is a rare honor, and one that highlights the the importance of Robins’ contributions to the fields of statistics and epidemiology. The special issue of Statistical Science is accessible online via Project Euclid: "Special Issue on Semiparametrics and Causal Inference" (Volume 29, Number 4).
