Accession Number : AD0709926

Title :   A GOODNESS OF FIT TEST FOR BIVARIATE NORMAL DISTRIBUTIONS.

Descriptive Note : Master's thesis,

Corporate Author : NAVAL POSTGRADUATE SCHOOL MONTEREY CALIF

Personal Author(s) : Miller,James Edward

Report Date : APR 1970

Pagination or Media Count : 39

Abstract : The paper is an investigation of a goodness of fit test for bivariate normal distributions. The test procedure is based on random linear functions of bivariate normal random variables. The test makes use of the maximum Kolmogorov D(M) statistic over the linear functions which are computed. An estimate of the distribution of M is obtained by computer simulation. No attempt is made to determine the power of the test. (Author)

Descriptors :   (*STATISTICAL DISTRIBUTIONS, MULTIVARIATE ANALYSIS), (*STATISTICAL TESTS, CURVE FITTING), COMPUTER PROGRAMS, RANDOM VARIABLES, SAMPLING, THESES

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE