Generation of Multivariate Random Variables with Known Marginal Distribution and a Specified Correlation Matrix

by Isidro R. Cruz_Medina, Mucio Osorio_Sánchez and Fernando García_Páez.

Abstract: An algorithm for generating correlated random variables with any set of marginal continuous distributions and a specified correlation matrix is provided. An example for generating a sample from a trivariate distribution with gamma marginals and positive and negative correlations is provided.

Key Words: Random vector generation, Multivariate distributions, Bivariate gamma distribution, Bivariate beta distribution

Authors:
Isidro R. Cruz_Medina, rcruz@itson.mx
Mucio Osorio_Sánchez, mosorio@itson.mx
Fernando García_Páez, garpaes@uas.uasnet.mx

Editor: Ravindra Khattree, khattree@oakland.edu

READING THE ARTICLE: You can read the article in portable document (.pdf) format (54716 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2208 times since JUNE 14, 2010.

Return to the InterStat Home Page.