Seminar: Dipak Dey

Title

The Complex Elliptical Distributions with Applications to Shape Analysis 

Speaker

Dipak Dey, Chair, Department of Statistics, University of Connecticut

Abstract

We develop a general class of complex elliptical distributions and extend it on a complex sphere. Such class includes many distributions, eg., complex Watson, Bingham, angular central Gaussian and several others. We study some properties of this class of distributions and apply the distribution theory for modeling shapes in two dimension. Maximum likelihood and Bayesian methods of estimation are developed. Using Markov chain Monte Carlo method, credible regions for shapes are also developed. The methodology is exemplified through an example on estimation of shape of mouse vertebra.