Event Calendar

Share This

MHRI-GHUCCTS Monthly Statistical Seminar: Regression Models

Date Fri, Nov 20
Time 12: 00 PM - 1: 00 PM
Location Online

This month's MHRI-GHUCCTS Monthly Statistical Seminar Series features Dr. Paul Kolm, Associate Director in the Department of Biostatistics and Biomedical Informatics at MedStar Health Research Institute. Dr. Kolm will discuss regression models.

Abstract: Many of the methods for developing multivariable regression models that commonly appear in the research literature are not valid.  More recent methods such as elastic-net and random forests avoid the problems of older methods and accommodate assessment of a large number of variables that potentially predict or are associated with the outcome of interest.  In this session, we will review both older and more recent methods for developing multivariable regression models. 

Dr. Kolm is Associate Director, Department of Biostatistics and Biomedical Informatics at MedStar Health Research Institute. He has over 30 years of experience in consulting with principal investigators in the design and analysis of clinical trials, retrospective and observational studies, and large patient registries. He has served as Statistical Editor of JACC-Interventions, served on the Scientific and Clinical Support Committee of the American College of Cardiology National Cardiovascular Data Registry (NCDR), and on the American Heart Association Epidemiology, Prevention, Outcomes and Behavioral study section. While at Emory University, he was the lead biostatistician for the Grady Hospital General Clinical Research Center (GCRC) and a member of the GCRC Advisory Committee, which reviewed Emory and Grady research protocols for scientific merit. He is currently the lead biostatistician for several NIH and industry-funded studies. Dr. Kolm’s statistical areas of interest include cost-effectiveness analyses, multiple imputation methods for missing data, analysis of sparse outcomes and multivariable regression models where the number of variables is greater than the number of participants.