THE GENERALIZED LINEAR MIXED MODEL LEADING TERMS

Relatore: prof. Matt Wand - University of Technology Sydney, Australia

Abstract
Generalized linear mixed models were born in the early 1990s as
the love child of linear mixed models (1950s) and generalized 
linear models (1970s). Now, in the 2020s, every day ends with the 
publication of around 3 new papers on the topic. Despite their 
ever-increasing ubiquity, there has been very little in the way of 
asymptotic theory for the maximum likelihood estimators of generalized 
linear mixed model parameters. Apart from simple conveyance of estimator 
behaviour, there are the usual payoffs concerning statistical inference,
sample size calculations and optimal design. This talk will describe
new results concerning the generalized linear mixed model leading terms, and 
is joint with Jiming Jiang, Aishwarya Bhaskaran and Luca Maestrini.
Ramifications concerning variational approximation will also be mentioned. 

SHORT BIOGRAPHY

Matt Wand is a Distinguished Professor of Statistics at University
of Technology Sydney, Australia. He is an elected fellow of the American
Statistical Association, the Institute of Mathematical Statistics and
the Australian Academy of Science. His main research areas are generalized
linear mixed models, semiparametric regression and variational approximate 

inference.