Characterizing variation of nonparametric random probability measures using the Kullback–Leibler divergence. Issue 3 (4th May 2017)

• Record Type: Journal Article
• Authors: Watson, J.; Nieto-Barajas, L.; Holmes, C.
• Is Part Of: Statistics. Volume 51:Issue 3(2017)
• Page Start: 558
• Page End: 571
• Publication Date: 2017-05-04
• Subjects: Bayesian nonparametrics -- Kullback–Leibler divergence -- bootstrap methods -- Pólyatrees; Mathematical statistics -- Periodicals; 519.505
• DOI: 10.1080/02331888.2016.1258072
• Languages: English
• ISSNs: 0233-1888

Abstract

ABSTRACT: This work characterizes the dispersion of some popular random probability measures, including the bootstrap, the Bayesian bootstrap, and the Pólya tree prior. This dispersion is measured in terms of the variation of the Kullback–Leibler divergence of a random draw from the process to that of its baseline centring measure. By providing a quantitative expression of this dispersion around the baseline distribution, our work provides insight for comparing different parameterizations of the models and for the setting of prior parameters in applied Bayesian settings. This highlights some limitations of the existing canonical choice of parameter settings in the Pólya tree process.