A Data Based Random Number Generator for a Multivariate Distribution - A Users' Manual
Abstract
Let X be a k-dimensional random variable serving as input for a system with output Y(not necessarily of dimension k). Given X, an outcome Y or a distribution of outcomes G(Y/X) may be obtained either explicitly or implicitly. We consider here the situation in which we have a real world data set X (J) to the nth power (j=1) to the n power and a means of simulating an outcome Y. A method for empirical random number generation based on the sample of observations of the random variable X without estimating the underlying density is discussed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1982
- Accession Number
- ADA121571
Entities
People
- Barry A. Bodt
- Malcolm S. Taylor
Organizations
- Ballistic Research Laboratory