Testing Hypotheses in Linear Models with Weighted Rank Statistics.
Abstract
Tests of hypotheses for the parameters in a general linear model are considered based on weighted rank statistics. Results are presented for tests based on a rank estimate, tests based on drop in dispersion and aligned rank tests. Weights can be used to focus the analysis on simple effects and provide an additional degree of robustness to rank tests. Several analysis of variance applications are discussed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1983
- Accession Number
- ADA135759
Entities
People
- G. L. Sievers
Organizations
- Western Michigan University