This award supports travel for participants in the symposium Advances in Statistical Methods for the Analysis of Observational and Experimental Data, held July 13, 2013 on the North Carolina State University campus in Raleigh, North Carolina. The past several decades have seen numerous significant advances in statistical theory and methodology for the analysis of data from experimental and observational studies. These developments have had a profound impact on statistical practice. The conference features presentations by leaders in the development of statistical theory and methods in areas such as censored time-to-event data, missing data, longitudinal data, and causal inference.
The conference will serve as a forum for discussion of developments in statistical methodology and theory that have propelled advances in research in the health, biological, epidemiological, social, and economic sciences. The conference offers a perfect opportunity for junior researchers to gain a valuable perspective on emerging trends and the opportunity to interact with senior and mid-career level leaders responsible for past and present advances.
Conference web site: www.stat.ncsu.edu/events/2013_tsiatis_symposium/
The symposium provided an outstanding program that offered participants, including senior and junior researchers, statistical practitioners, and students, the opportunity to hear the perspectives of leaders in the field through a series of invited lectures highlighting past, present, and future directions. Participants were also able to present their own work during a dedicated contributed poster session.The conference was broadly about the analysis of data from experimental and observational studies, often from clinical trials and other medical studies. There were 138 registered particpants, and the talks were well-attended. The attached images are of Jamie Robins giving a talk and NC State graduate student Nick Meyer explaining his poster to several participants. The past several decades have seen numerous significant advances in statistical theory and methodology for the analysis of data from experimental and observational studies. These developments have had a profound impact on statistical practice. Methods for the analysis of censored, time-to-event data have revolutionized studies of reliability in engineering and of the progression of disease. Constructing formal structures for making causal inference in the presence of the confounding inherent in observational data has led to new insights in epidemiologic, economic, sociological, behavioral, and other contexts and to innovative new theory, particularly when the phenomena under study are time-dependent. Even the most meticulously planned experimental study can be plagued by missing data, data that were intended to be collected but werent, and important measurements can be obscured by device or assay error. Methods for handling these complications have become broadly accepted and used in a variety of disciplinary areas. The longitudinal nature of many studies across all areas of scientific inquiry has led to sophisticated theory and methods for analysis that are now commonplace. The theory of semiparametrics applied to these settings has led to a deeper understanding of the structure of many statistical problems along with elegant, widely-used methodology.
