Statistics Seminar: Tereza Neocleous

Tereza Neocleous, University of Glasgow

Friday 31 May 2013, 1400-1500
A54, Postgraduate Statistics Centre Lecture Theatre

A Bayesian hierarchical model for compositional data

Compositional data frequently occur in chemistry. One example is data points corresponding to the percentage with respect to weight of each chemical element present in a glass sample. Such compositions are subject to a sum constraint which should be taken into account when the data are analysed. In addition, it is not unusual in such elemental compositions to observe zero concentrations for some of the components. We present an approach to dealing with zeros in glass elemental compositions, and develop a Bayesian hierarchical model for glass data with the purpose of classifying glass for forensic purposes and for assessing the evidential value of glass as transfer evidence.