DID YOU KNOW? DTIC has over 3.5 million final reports on DoD funded research, development, test, and evaluation activities available to our registered users. Click HERE
to register or log in.
CONTRIBUTIONS TO THE THEORY OF NUMERICAL SIMULATION.
TEXAS UNIV AT AUSTIN LABS FOR ELECTRONICS AND RELATED SCIENCE RESEARCH
Recently there has been interest in the pattern recognition literature in multivariate generalizations of the Pearson Type 2 and Pearson Type 7 distributions. As yet methods for generating a random sample from these distributions have not appeared. Further, little has been done toward generating a random sample from the Wishart distribution. This report is aimed at providing efficient techniques for producing random samples of these types. Methods and supporting theory for generating random samples from these distributions are presented and discussed. Finally, two general approaches to problems in simulation of this general type are developed.