Robustness of Yao’s, James’, and Johansen’s Tests Under Variance-Covariance Heteroscedasticity and Nonnormality

James Algina, Takako C. Oshima and K. Linda Tang

Journal of Educational and Behavioral Statistics, 1991, vol. 16, issue 2, 125-139

Abstract: Type I error rates for Yao’s, James’ first order, James’ second order, and Johansen’s tests of equality of mean vectors for two independent samples were estimated for various conditions defined by the degree of heteroscedasticity and nonnormality (uniform, Laplace, t(5), beta (5, 1.5), exponential, and lognormal distributions). For these alternatives to Hotelling’s T 2 , variance-covariance homogeneity is not an assumption. Although the four procedures can be seriously nonrobust with exponential and lognormal distributions, they were fairly robust with the rest of the distributions. The performance of Yao’s test, James’ second order test, and Johansen’s test was slightly superior to the performance of James’ first order test.

Keywords: multivariate analysis; Behrens-Fisher problem; Welch-James procedure (search for similar items in EconPapers)
Date: 1991
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.sagepub.com/doi/10.3102/10769986016002125 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sae:jedbes:v:16:y:1991:i:2:p:125-139

DOI: 10.3102/10769986016002125

Access Statistics for this article

More articles in Journal of Educational and Behavioral Statistics
Bibliographic data for series maintained by SAGE Publications ().