Professor Peter T Kim

Department of Mathematics and Statistics

University of Guelph

Title: " Wishart Mixtures Diffusion Tensor Imaging and the Statistics of Harish-Chandra "

Abstract:
To say the least, the work of Harish-Chandra has had profound consequences
in the development of mathematics. To some degree, the work is also apparent
in statistics, however, due to the fact that most of statistics take place in
a commutative setting, the profound non-commutative results are somewhat hidden.
In this talk, we will assume non-commutativity, and by doing so, we will
demonstrate the contribution to statistics in the setting of Wishart mixture
models on the space of positive definite symmetric matrices. This has practical
relevance to Diffusion Tensor Imaging which is a modern in vivo brain imaging
technique that tracts the whole brain white matter fibers.