Spring Seminar Series  May 1, 2008
University of Minnesota
School of Statistics
College
of Liberal Arts

Effects of Ignoring Correlations in Longitudinal Response or Covariate Processes

 

Naisyin Wang

Texas A&M University

Thursday, May 8, 2008
3:30 PM, 115 Ford Hall
Minneapolis, East Bank Campus
Social at 3:00 PM, 300 Ford Hall

 

Abstract

The analysis of hierarchical biomedical data sometimes requires more modeling flexibility than that can be provided 
by standard parametric approaches.  It is commonly believed that the effect of ignoring covariance structure is mainly 
on the lost of efficiency. In this talk, I will use some numerical outcomes and examples to illustrate some potential 
concerns when one ignores the correlations in longitudinal measurements. The less known fact is the serious level of 
biases that  could be induced by ignoring existing correlations. Some solutions would be provided.