
(a) W=TI{1/(l+cj)}, (b) T-=n nc3, (c) V=n n{cj/(l+cj)}. The test (a) is the likelihood ratio criterion (Wilks, 1932). Bartlett (1939) proposed (b) and (c) as reasonable alternatives to (a) for testing the general linear hypothesis. The test (b) was studied by Hotelling (1947, 1951) and (c) has been studied intensively by Pillai (1964, 1966). Asymptotic formulae have been proposed for the distribution functions of W and T2; see, for example, Sugiura & Fujikoshi (1969), Ito (1956, 1960), an unpublished report of M. Siotani, and Lee (1971). In this paper asymptotic formulae are derived for the distribution function of V in the general case and the percentage points for V in the null case. Using the asymptotic formulae, the powers of the three tests (a), (b) and (c) are compared numerically.
Multivariate distribution of statistics, Hypothesis testing in multivariate analysis
Multivariate distribution of statistics, Hypothesis testing in multivariate analysis
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 41 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Top 10% | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
