Seminars

General Departmental Seminar Series

Analysis of Clustered Data: A Combined Estimating Equations Approach

Julie Stoner, Ph.C.-University of Washington-Seattle

Tuesday, February 22, 2:00-3:00 p.m.

Conference Room, 504 N. Walnut

ABSTRACT

Clustered data often arise in biomedical research. Examples include data from longitudinal studies and data sampled within clusters. In this seminar, I will propose a new regression analysis method for clustered data that optimally weights and combines sources of information from the data by optimally combining estimating equations. The proposed method also avoids modeling decisions regarding the true correlation structure of the data and in particular settings, results in increased estimation efficiency relative to generalized estimating equations. An example, in which between- and within-cluster information is optimally combined, will be presented. Increased estimation efficiency of the proposed method, relative to a generalized estimating equations approach, will be demonstrated through derivations and simulations. An application of the methodology will also be discussed.

Back to General Departmental Seminar Series