Brunner, Lawrence J.; Lo, Albert Y. Bayes methods for a symmetric unimodal density and its mode. (English) Zbl 0697.62003 Ann. Stat. 17, No. 4, 1550-1566 (1989). The main problem considered here is the estimation of an unknown symmetric unimodal density and its mode, though attention is also paid to the related question of the estimation of a decreasing density on the half-line. A Bayes solution is obtained, using a Dirichlet process prior as the mixing distribution. Posterior quantities turn out to be finite sums, whose evaluation is effected by using Monte Carlo methods based on sampling finite Markov chains. Reviewer: A.Dale Cited in 1 ReviewCited in 36 Documents MSC: 62A01 Foundations and philosophical topics in statistics 62F15 Bayesian inference 62G05 Nonparametric estimation Keywords:simulation of Markov chains; mixture models; symmetric unimodal density; Dirichlet process prior; mixing distribution; Monte Carlo methods PDFBibTeX XMLCite \textit{L. J. Brunner} and \textit{A. Y. Lo}, Ann. Stat. 17, No. 4, 1550--1566 (1989; Zbl 0697.62003) Full Text: DOI