A Bivariate Test of Goodness of Fit Based on a Gradually Increasing Number of Order Statistics
Abstract
An important area of Statistical Inference relates to the problem of assessing the conformity or goodness of fit of some observations to a null hypothesis. This hypothesis may be that the observations came from a distribution which belongs to a given family or may be a completely specified distribution. A statistical procedure which solves such a problem is called a test of goodness of fit. This thesis is concerned with the problem of testing the goodness of fit of a sample drawn from a continuous bivariate distribution, where the null hypothesis is that the true distribution is a completely specified one. A test is developed, where the test criterion is based in some functions of a subset of order statistics, functions which depend on the distribution under the null hypothesis.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1975
- Accession Number
- ADA008205
Entities
People
- Jose Kreimerman
Organizations
- Cornell University