Random variate generation for multivariate unimodal densities
- 1 October 1997
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Modeling and Computer Simulation
- Vol. 7 (4) , 447-477
- https://doi.org/10.1145/268403.268413
Abstract
A probability density on a finite-dimensional Euclidean space is orthounimodal with a given mode if within each orthant (quadrant) defined by the mode, the density is a monotone function of each of its arguments individually. Up to a linear transformation, most of the commonly used random vectors possess orthounimodal densities. To generate a random vector from a given orthounimodal density, several general-purpose algorithms are presented; and an experimental performance evaluation illustrates the potential efficiency increases that can be achieved by these algorithms versus naive rejection.Keywords
This publication has 24 references indexed in Scilit:
- Bivariate Distributions with Exponential ConditionalsJournal of the American Statistical Association, 1988
- A new fast method for computer generation of gamma and beta random variables by transformations of uniform variablesStatistics, 1987
- Methods for generating random variates with Polya characteristic functionsStatistics & Probability Letters, 1984
- Algorithm 599ACM Transactions on Mathematical Software, 1983
- A Class of Multivariate DistributionsAustralian Journal of Statistics, 1981
- Gamma variate generators with increased shape parameter rangeCommunications of the ACM, 1980
- Some Simple Gamma Variate GeneratorsJournal of the Royal Statistical Society Series C: Applied Statistics, 1979
- The Generation of Gamma Variables with Non-Integral Shape ParameterJournal of the Royal Statistical Society Series C: Applied Statistics, 1977
- Symmetric random walkTransactions of the American Mathematical Society, 1962
- Bivariate Exponential DistributionsJournal of the American Statistical Association, 1960