Murray has been a Senior Lecturer with the Department since 1986. Previously he held consulting or academic positions with the New Zealand Ministry of Agriculture and Fisheries, the University of Botswana and Wilfrid Laurier
University, Ontario, Canada. His PhD from the University of British Columbia is in Pure Mathematics.
Biometrics; robust estimation and diagnostics; mixture models; statistical computing.
Murray's current research is in the area of statistical computing, especially
in applications of the EM algorithm. Mixtures of statistical models is an example
of such an application and can be quite useful as many real data sets can profitably
be regarded as mixtures from several sources. Implementing the Minimum
Message Length approach to estimation and model selection using the EM
algorithm has been a recent preoccupation. Lately he has been working with large data
sets obtained from the internet. Internet data ranges over many orders of magnitude
and has many interesting features.
Multimix: a model-based clustering program
Multimix ftp site contains a Fortran 77 program for fitting a class of models that includes latent class models and finite mixtures of multivariate normal distributions. An allocation rule based on the fitted model can be used as a form of cluster analysis. Documentation and a background paper is included at the ftp site.