Tests on multiple correlation coefficient and multiple partial correlation coefficient
Let X1(1 - 1), X2(1 - p2), and X3(1 - p3) be three sets of random variables distributed jointly as a normal distribution. Let [varrho]1.23 and [varrho]1.2 be the multiple correlation coefficients between X1 and (X2, X3) and between X1 and X2, respectively. Invariant tests for the following four testing problems are considered: [varrho]1.23 = 0 vs [varrho]1.23 > 0, [varrho]1.23 = 0, vs [varrho]1.23 > 0, [varrho]1.23 = [varrho]1.2 vs [varrho]1.23 > [varrho]1.2, [varrho]1.2 = 0 vs [varrho]1.2 > 0. The joint distribution of the sample multiple correlation coefficients R1.23 and R1.2 is derived.
Year of publication: |
1977
|
---|---|
Authors: | Gupta, Somesh Das |
Published in: |
Journal of Multivariate Analysis. - Elsevier, ISSN 0047-259X. - Vol. 7.1977, 1, p. 82-88
|
Publisher: |
Elsevier |
Keywords: | Multiple correlation multiple partial correlation invariant tests |
Saved in:
Saved in favorites
Similar items by person
-
Monotonicity and unbiasedness properties of ANOVA and MANOVA tests
Gupta, Somesh Das, (1980)
-
Brunn-Minkowski inequality and its aftermath
Gupta, Somesh Das, (1980)
-
On the power of Wilks' U-test for MANOVA
Gupta, Somesh Das, (1973)
- More ...