Accession Number : AD0704325

Title :   TRANSFORMATION OF A SET OF PSEUDO-RANDOM NUMBERS INTO A SET REPRESENTING ANY DESIRED PROBABILITY AND CORRELATION.

Descriptive Note : Technical rept.,

Corporate Author : ARMY MISSILE COMMAND REDSTONE ARSENAL ALA ADVANCED SENSORS LAB

Personal Author(s) : Holliday,Edward M.

Report Date : 12 DEC 1969

Pagination or Media Count : 160

Abstract : Many problems in engineering, economics, mathematics, and the sciences require investigations using digital simulations. Frequently the simulation of stochastic processes is required and is performed by a computer program which generates a pseudo-random sequence of numbers to represent random samples of the process. In general, it is not possible to control the probability and the correlation of the sequence at will. The objective of this report is the development of a general technique for digital generation of a set representing noise with independent control of the probability and t9 For this purpose, the set transformation method is developed and implemented in the computer subroutine identified as XFORM. A set of pseudo-random numbers is transformed into a second set which represents uncorrelated or correlated samples of any probability distribution. Existing methods of generating pseudo-random numbers representing uniform and nonuniform probability distributions, both uncorrelated and correlated, are reviewed. Two procedures for introducing correlation are developed. Two subroutines are developed to analyze random sets.

Descriptors :   (*STOCHASTIC PROCESSES, MATHEMATICAL MODELS), (*STATISTICAL DISTRIBUTIONS, SAMPLING), CORRELATION TECHNIQUES, MONTE CARLO METHOD, PROBABILITY DENSITY FUNCTIONS, MODEL THEORY, SIMULATION, SUBROUTINES

Subject Categories : Statistics and Probability
      Operations Research
      Computer Programming and Software

Distribution Statement : APPROVED FOR PUBLIC RELEASE