SEMINARS
|
|
Fall 2007
STATISTICS
COLLOQUIUM
Wednesday, October 31, 2007
3:30-4:00—Refreshments--Yost 327
4:00-5:00—Talk--Yost 101
Vinay Cheruvu, PhD Candidate
Department of Epidemiology and Biostatistics
Case Western Reserve University
Mediation and Causal Inference
Mediation is a causal chain that explains the causal pathway between an exposure and an outcome by acting as a third variable. This talk introduces the formal definition of a mediation effect and introduces classical approaches to estimate a mediation effect. An attempt is made to distinguish between a mediator and a confounder. Focus would be on how to estimate the mediation effect when the mediator is binary and introducing longitudinal mediation.
|
