Lectures and seminars Sfo Epi Seminar in Biostatistics: Professor Thomas Lumley
Welcome to attend the SfoEpi Seminar Series in Biostatistics!
Title: Pairwise composite likelihood for mixed models under complex sampling
Speaker: Professor Thomas Lumley, Chair in Biostatistics at The University of Auckland, New Zealand
Zoom Link: https://ki-se.zoom.us/j/67206843430
*Webinar format: unless you are on the panel you will be muted and with no video. Participants can raise their hands during the Q&A to be recognized to ask questions.
Abstract
Most (frequentist) exploratory and inferential analyses can now be adjusted for complex sampling, so that the inferences are to the source population and the questions do not depend on how the sampling was done. The most important exception is mixed models, where the current standard approach requires the sampling units to be the model clusters, and even so does not produce unbiased estimating equations. Pairwise composite likelihood, however, provides a way to fit mixed models to very general designs. Our motivating example was fitting linear mixed models with ancestry and relatedness terms to data from the Hispanic Community Health Survey/Study of Latinos. I will describe how the method works and present some simulations. I will also explain why we still cannot handle the motivating example, but have hopes to do so soon.
If you have any questions, please contact Erin Gabriel at Erin.Gabriel@ki.se