Department of
presents
Dr. Annie
Qu
Model
Diagnostic Tests for Selecting Informative
Correlation
Abstract
In the generalized method of moments approach
to longitudinal data analysis, unbiased estimating functions can be
constructed to incorporate both the marginal mean and the correlation
structure of the data. Increasing the number of parameters in the correlation
structure corresponds to increasing the number of estimating functions. Thus,
building a correlation model is equivalent to selecting estimating functions.
This paper proposes a chi-squared test to choose informative unbiased
estimating functions. We show that this methodology is useful to identify
which source of correlation is important to incorporate when there are
multiple possible sources of correlation. This method can also be applied to
determine the optimal working correlation for the generalized estimating
equation approach. Lung cancer prevention data from MD Anderson will be
illustrated for the proposed method.
This is joint work with J. Jack Lee of MD
Friday,
November 30, 2007
3:35 - 4:35 pm
301 Riddick Hall
Refreshments will be served in the common area of 301 Riddick at 3:00
pm.