Edgeworth expansion of Wilks' lambda statistic
An asymptotic expansion of the null distribution of the Wilks' lambda statistic is derived when some of the parameters are large. Cornish-Fisher expansions of the upper percent points are also obtained. A monotone transformation which reduces the third and the fourth order cumulants is also derived. In order to study the accuracy of the approximation formulas, some numerical experiments are done, with comparing to the classical expansions when only the sample size tends to infinity.
Year of publication: |
2006
|
---|---|
Authors: | Wakaki, Hirofumi |
Published in: |
Journal of Multivariate Analysis. - Elsevier, ISSN 0047-259X. - Vol. 97.2006, 9, p. 1958-1964
|
Publisher: |
Elsevier |
Keywords: | MANOVA High dimension Likelihood ratio Monotone transformation Edgeworth expansion |
Saved in:
Saved in favorites
Similar items by person
-
Wakaki, Hirofumi, (1998)
-
Fujikoshi, Yasunori, (2005)
-
Simple Formula for Calculating Bias-corrected AIC in Generalized Linear Models
Imori, Shinpei, (2014)
- More ...