🚀 go-pugleaf

Colloquium, University of Toronto, Dept. of Statistics

Topic:   Generalized Linear Models with Dependent Responses
Speaker: Scott L. Zeger, Johns Hopkins University
Date:    Thursday, December 7, 1989   4:00 - 5:00
Place:   Room 1085, Sidney Smith Hall, 100 St George Street, U of T
Abstract:
     Generalized linear models have unified the approach to regression for
a wide variety of discrete, continuous and censored response variables which
can be assumed to be independent across experimental units.  In applications 
such as longitudinal studies, genetic studies of families and survey sampling,
observations may be obtained in clusters.  Responses from the same cluster can
not be assumed to be independent.  With linear models, correlation has been
effectively modelled by assuming there are cluster-specific coefficients
(random effects) which derive from an underlying mixing distribution.
Extensions of generalized linear models to include random effects has
thus far been hampered by the need for higher order integration to evaluate
likelihoods.

     In this talk, we cast the generalized linear random effects model in
a Bayesian framework and use a recent Monte Carlo method, the Gibb's
sampler, to overcome the current numerical limitations.  The resulting 
algorithm is flexible to easily accommodate changes in the number of random
effects and in their assumed distribution.  The methodology is illustrated
through an analysis of infectious disease data.
-------

Coffee and tea will be served in the De Lury Lounge (SS6006) at 3:30 p.m.