A service provided by the WU Library and the WU IT-Services

Transformed Density Rejection with Inflection Points

Botts, Carsten and Hörmann, Wolfgang and Leydold, Josef (2011) Transformed Density Rejection with Inflection Points. Research Report Series / Department of Statistics and Mathematics, 110. WU Vienna University of Economics and Business, Vienna.

PDF (Manuscript)
Download (199Kb) | Preview
[img] Plain Text (Description of Electronic Supplement)
Download (422b)
[img] Plain Text (Example 1: Bimodular distribution)
Download (1208b)
[img] Plain Text (Example 2: Exponential power distribution)
Download (2641b)
[img] Plain Text (Example 3: Generalized Inverse Gaussian distribution)
Download (2535b)
[img] Plain Text (R code for proof-of-concept implementation of Algorithm Tinflex-log)
Download (21Kb)
[img] Archive (TGZ) (R package Tinflex for most general algorithm)
Download (20Kb)


The acceptance-rejection algorithm is often used to sample from non-standard distributions. For this algorithm to be efficient, however, the user has to create a hat function that majorizes and closely matches the density of the distribution to be sampled from. There are many methods for automatically creating such hat functions, but these methods require that the user transforms the density so that she knows the exact location of the transformed density's inflection points. In this paper, we propose an acceptancerejection algorithm which obviates this need and can thus be used to sample from a larger class of distributions.

Item Type: Paper
Keywords: nonuniform random variate generation / transformed density rejection / inflection points
Classification Codes: MSC 65C05, 65C10
Divisions: Departments > Finance, Accounting and Statistics > Statistics and Mathematics
Depositing User: Josef Leydold
Date Deposited: 18 Jul 2011 11:24
Last Modified: 15 Sep 2017 10:54
FIDES Link: https://bach.wu.ac.at/d/research/results/55154/
URI: http://epub.wu.ac.at/id/eprint/3158


View Item