SCALE MIXTURES DISTRIBUTIONS, BAYESIAN ROBUSTNESS AND MCMC

  • Speaker: Dr Boris Choy, Department of Mathematical Sciences, University of Technology, Sydney
  • Time: 4:00p.m. Wednesday 20th October 2004
  • Venue: Red Centre Building Room RC-3084
  • near Barker Street Gate 14

This talk presents two classes of scale mixtures distributions
which provide a good alternative to statistical modelling when the
normal distribution is inappropriate. We shall investigate the
contributions of these classes of distributions in Bayesian robustness.
Tecnhical details of incorporating the distributions in the Markov
chain Monte Carlo algorthithms will be discussed. For illustrative
purpose, a financial time series model is analysed using these two
classes of distributions.