SKEW-T AND ALL THAT: Some of the main parametric families of distributions and how they compare
Title: SKEW-T AND ALL THAT: Some of the main parametric families of distributions and how they compare
Speaker: Prof. Chris Jones, Open University.
Date: Thursday, 8th October 2015
Time: 4pm – 5pm
Location: Room E0.01, Science East
Abstract:
Univariate continuous distributions are one of the fundamental components on which statistical modelling, ancient and modern, frequentist and Bayesian, multidimensional and complex, is based.
In this talk, I will review and compare some of the main general techniques for providing families of typically unimodal distributions on R with one or two shape parameters, controlling skewness and/or tailweight, in addition to their all-important location and scale parameters.
One important and useful family is comprised of the skew-symmetric distributions brought to prominence by Azzalini. As these are widely covered in the literature, I focus more on their complements and competitors. Principal amongst these are distributions formed by transforming random variables, by what I call transformation of scale, including two-piece distributions, and by probability integral transformation of non-uniform random variables.
In the last part of the talk, I shall look more specifically at some skew-t distributions arising from these general considerations, making further comments and (theoretical) comparisons as I go.
Series: Statistics
Social Media Links