American Statistical Association
There is a long-standing interest in the evaluation of gene-gene and gene-environment interactions in genetic and epidemiology studies. Evaluation of interactions focuses on the structure of the expected response (or their contrasts) in the various risk factors sub-classes. Once the outcome is modeled as a function of a given set of risk factors, interaction is generally defined as a departure (of the model) from additivity on a certain scale in which the data are measured. The distribution of the outcome under two risk factors may have the same mean, but different variances, which may also be taken as evidence for an interaction between the two risk factors. This work examines these concepts and investigates the benefits and limitations of data transformations for obtaining further insights into these concepts when the data are unbalanced and the trait is binary. Data on smoking and NAT2 acetylation from two published cancer studies are used as illustrative examples.
|Date:||Wednesday, November 9, 2011|
|Time:||4:00 - 5:00 P.M.|
Memorial Sloan-Kettering Cancer Center
Department of Epidemiology and Biostatistics
307 East 63rd Street
(between First and Second Avenues)
New York, New York
Note: To gain access to the building, please follow the directions by the telephone in the foyer.