On the admissibility and consistency of tests for homogeneity of variances (Q582747)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | On the admissibility and consistency of tests for homogeneity of variances |
scientific article |
Statements
On the admissibility and consistency of tests for homogeneity of variances (English)
0 references
1989
0 references
The authors consider the hypothesis testing problem for the homogeneity of variances in the one-way fixed and balanced analysis of variance model under the assumptions of independence and normality. They obtain a necessary and sufficient condition for admissibility of a test among the class of scale invariant tests. In particular they show that some classical tests such as Hartley's test and Cochran's test are not admissible. A certain limiting property (parameter consistency) of scale invariant tests is defined and it is shown that several well-known tests are parameter consistent for all significance levels. They also show that some well-known tests may not be parameter consistent or may be parameter consistent only for certain significance levels. Extensions to non-normal cases are also indicated.
0 references
maximal invariant parameter
0 references
Kullback-Leibler distance
0 references
likelihood ratio test
0 references
locally most powerful unbiased test
0 references
robust tests
0 references
homogeneity of variances
0 references
one-way fixed and balanced analysis of variance model
0 references
independence
0 references
normality
0 references
class of scale invariant tests
0 references
Hartley's test
0 references
Cochran's test
0 references
parameter consistency
0 references
significance levels
0 references
non-normal cases
0 references