UCI Applied Statistics Seminar

Tuesday, November 21, 2000
Noon - 1:00 p.m.
Social Science Plaza B-1222


Geoff McLachlan

Professor of Statistics
Department of Mathematics
University of Queensland
St. Lucia, Brisbane, Australia


Applying Mixture Models to Data

In recent years, there has been considerable attention given to the use of finite mixture distributions to model data on a variety of random phenomena. In this talk, we consider some of these applications with discussion of some of the computational issues arising with the fitting of mixture models via the so-called expectation--mazimization algorithm. The use of the EMMIX software for these purposes is to be demonstrated. We also consider the role of mixture models in survival analysis.


About the Speaker

Educational History
1969 BSc (Hons.) University of Queensland
1973 PhD (Mathematical Statistics) University of Queensland
1994 DSc (Mathematical Statistics) University of Queensland

Research Interests
McLachlan's research in statistics is in the related fields of classification, cluster and discriminant analyses, image analysis, machine learning, neural networks, and pattern recognition, and in the field of statistical inference. His focus in the latter field has been on the theory and applications of finite mixture models and on estimation via the EM algorithm. A common theme of his research in these fields has been statistical computation, with particular attention being given to the computational aspects of the statistical methodology. This computational theme extends to his more recent interests in the field of data mining. He is also actively involved in research in the field of medical statistics.

Books
The speaker's list of publications includes numerous research articles and four monographs. His third monograph on the EM algorithm (with T. Krishnan) received the Wiley-Interscience outstanding book award for 1997. A fourth monograph on finite mixture models (with David Peel) has just been published.


For further information about this event, please contact
The UCI Center for Statistical Consulting
[949] 824-1680