THE CHI-SQUARE TEST OF GOODNESS OF FIT FOR A MULTIVARIATE NORMAL DISTRIBUTION.
Abstract
The Pearson Chi-Square Goodness of Fit Criterion for testing the null hypothesis 'The random sample is from a univariate normal parent population' is generalized for multivariate normal distributions to test the null hypothesis 'The random sample of N-dimensional observation vectors is from an N-dimensional multivariate normal parent population.' The basis of the formulation is the distribution of the quadratic form in the exponent of the hypothesized distribution. Application of the procedure is illustrated with a numerical example of experimental data from a weapon effectiveness study. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1967
- Accession Number
- AD0655628
Entities
People
- Carl B. Bates
Organizations
- Naval Surface Warfare Center Dahlgren Division