1997 Annual Meeting of the SSC in Fredericton

BIOSTATISTICS WORKSHOP
Generalized Linear Mixed Models
June 1, 1997
Fredericton, New Brunswick

 

A one-day workshop, sponsored by the Biostatistics Section, will be given by Charles E. McCulloch of Cornell University.

Generalized linear mixed models (GLMMs) are a natural outgrowth of both linear mixed models and generalized linear models (GLMs). As such, they are of wide applicability and practical importance. GLMMs can accommodate non-normally distributed responses, specification of a possibly non-linear link between the mean of the response and the predictors, and can model overdispersion and correlation by incorporating random effects.

The workshop will begin with introductory examples and motivation for the use of Generalised Linear Mixed Models. Both linear mixed models and Generalised Linear Models will be briefly discussed and then it will be shown how Generalised Linear Models can be extended through the inclusion of random factors. Models which can accommodate overdispersion and correlation without using random effects will be also briefly considered.

Methods of estimation and testing based on maximum likelihood, generalized estimating equations, penalized quasi-likelihood and Laplace approximations will be explained and contrasted. The availability and utility of some software for Generalised Linear Mixed Models will be described. Two case studies will be used to illustrate application of Generalised Linear Mixed Models. Current research topics in Generalised Linear Mixed Models will be briefly mentioned.

Registration fees:

• Regular - before April 15 $110 - after April 15 $135
  Student or Retired $35

To register , please make the appropriate choices on the enclosed (as a separate insertion) Conference Registration form.

For more information, please contact the organiser of the workshop

John Petkau
Department of Statistics
University of British Columbia
Vancouver, BC V6T 1Z2

john@stat.ubc.ca