Find us on Facebook Find us on LinkedIn Follow us on Twitter Subscribe to our YouTube channel

Business Analytics Seminars

The seminars are on Fridays at 11am in Room 498, Merewether Building (cnr of City Road and Butlin Avenue), unless otherwise specified.

The seminar organiser is Artem Prokhorov.

Upcoming Seminars

20th Feb 2015 - 11:00 am


Professor Geoff McLachlan,


Department of Mathematics; University of Queensland


Seminar Room 498, Merewether Building H04


Some Recent Results on Skew Versions of the Normal Distribution for Modelling Asymmetric Data with Applications to Financial Data


Normal and t-mixture models have been widely applied to datasets from a variety of fields, including biostatistics, bioinformatics, economics, finance, and image analysis, among many others. Finite mixtures of multivariate skew-symmetric distributions, in particular, the skew normal and skew t-mixture models, are emerging as a promising extension to the traditional normal and mixture modelling.  Most of these parametric families of skew distributions are closely related but they can vary widely in performance.  In this talk, we give a brief overview of various existing proposals for multivariate skew distributions. We compare the relative performance of some of these models in density estimation and clustering for two real datasets on some financial data.  The first set concerns some bankruptcy data on American firms.  The second set considers the estimation of the Value-at-Risk (VAR) from data on a portfolio of shares listed on the Australian Stock Exchange.