Computational Aspects of Optional Pólya Tree. Issue 1 (2nd January 2016)

• Record Type: Journal Article
• Authors: Jiang, Hui; Mu, John Chong; Yang, Kun; Du, Chao; Lu, Luo; Wong, Wing Hung
• Is Part Of: Journal of computational and graphical statistics. Volume 25:Issue 1(2016)
• Page Start: 301
• Page End: 320
• Publication Date: 2016-01-02
• Subjects: Bayesian nonparametrics -- Density estimation -- Recursive partition -- Smoothing -- Time complexity; Mathematical statistics -- Data processing -- Periodicals; Mathematical statistics -- Graphic methods -- Periodicals; 519.50285
• DOI: 10.1080/10618600.2014.1002927
• Languages: English
• ISSNs: 1061-8600

Abstract

Abstract : Optional Pólya tree (OPT) is a flexible nonparametric Bayesian prior for density estimation. Despite its merits, the computation for OPT inference is challenging. In this article, we present time complexity analysis for OPT inference and propose two algorithmic improvements. The first improvement, named limited-lookahead optional Pólya tree (LL-OPT), aims at accelerating the computation for OPT inference. The second improvement modifies the output of OPT or LL-OPT and produces a continuous piecewise linear density estimate. We demonstrate the performance of these two improvements using simulated and real date examples.