Summary
Srivastava [5] proposed a class of rank score tests for testing the hypothesis that β1=⋯β p =0 in the linear regression modely i =β1 x 1i +β2 x 2i +⋯+β p +x pi +ɛ i under weaker conditions than Hájek [2]. In this paper, under the same weak conditions, a class of rank score tests is proposed for testing β1=⋯β q =0 in the multivariate linear regression modely i =β1 x 1i +β2 x 2i +⋯+β p +x pi +ɛ i ,q≦p, where β i ’s arek-vectors. The limiting distribution of the test statistic is shown to be central χ 2 qk underH and non-central χ 2 qk under a sequence of alternatives tending to the hypothesis at a suitable rate.
Similar content being viewed by others
References
Hájek, J. (1961). Some extensions of the Wald-Wolfivitz-Noether theorem,Ann. Math. Statist.,32, 506–523.
Hájek, J. (1962). Asymptotically most powerful rank order tests,Ann. Math. Statist.,33, 1124–1147.
LeCam, L. (1960). Locally asymptotic normal families of distributions,Univ. of California Publication,3, No. 2, 37–98.
Srivastava, M. S. (1967). On fixed-width confidence bound for regression parameters and mean vector,J. R. Statist. Soc. B,29, 132–140.
Srivastava, M. S. (1968). On a class of non-parametric tests for regression parameters (abstract),Ann. Math. Statist.,39, 697.
Additional information
Research supported by Canada Council and National Research Council of Canada.
About this article
Cite this article
Srivastava, M.S. Asymptotically most powerful rank tests for regression parameters in manova. Ann Inst Stat Math 24, 285–297 (1972). https://doi.org/10.1007/BF02479758
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02479758