Forensic Speaker Comparison Using Evidence Interval in Full Bayesian Significance Test

Evidence and Credibility: Full Bayesian Significance Test for Precise Hypotheses

A Bayesian measure of evidence for precise hypotheses is presented. The intention is to give a Bayesian alternative to significance tests or, equivalently, to p-values. In fact, a set is defined in the parameter space and the posterior probability, its credibility, is evaluated. This set is the “Highest Posterior Density Region” that is “tangent” to the set that defines the null hypothesis. Our...

Full Bayesian Empirical Likelihood Significance Test for Equality Of Medians

This paper discusses the empirical likelihood ratio test for testing equality of medians of two distributions. Yu et al. (2011) proposed four forms of empirical likelihood ratio test for testing this hypothesis and one of them is based on the pooled sample median. The asymptotic distribution of the test statistic as derived in Yu et al (2011) differ from the one obtained in Nimbalkar and Rajars...

Full Bayesian significance test applied to multivariate normal structure models

The Full Bayesian Significance Test (FBST) for precise hypotheses is applied to a Multivariate Normal Structure (MNS) model. In the FBST we compute the evidence against the precise hypothesis. This evidence is the probability of the Highest Relative Surprise Set (HRSS) tangent to the sub-manifold (of the parameter space) that defines the null hypothesis. The MNS model we present appears when te...

Comparing Parameters of Two Bivariate Normal Distributions using the FBST: Full Bayesian Significance Test

Forensic speaker recognition: An evidence odyssey – Summary

The goal of a technology evaluation is to compare competing algorithms from a single technology. Testing of all algorithms is done on a standardized database collected by a “universal” sensor. Nonetheless, performance against this database will depend upon both the environment and the population in which it was collected. Because the database is fixed, results of technology tests are repeatable...