All-pairwise Multiple Comparison for Normal Mean Vectors Based on Tukey-Welsch's Procedure

Tsunehisa Imada


In this study we consider all-pairwise multiple comparison for several normal mean vectors. Specifically, intended to more powerful procedure compared to the single step procedure we apply Tukey-Welsch's step down procedure to our problem. We give some simulation results regarding critical values and power of the test intended to compare procedures.


Asymptotic distribution, Power of the test, Stepwise procedure

Full Text:



Anderson T W. An introduction to multivariate statistical analysis (Third ed.). Wiley, New York, 2003.

Tukey J W. The problem of multiple comparisons, Unpublished manuscript,

Princeton University, 1953.

Seo T, Siotani M. The multivariate Studentized range and its upper percentiles.

Journal of the Japan Statistical Society, 1992, 22: 123-137.

Fujikoshi Y, Seo T. Asymptotic expansions for the joint distribution of correlated Hotelling's statistics under normality. Communications in Statistics, Theory and Methods, 1999, 28: 773-788,.

Welsch R E. A modification of the Newman-Keuls procedure for multiple comparisons, Working Paper Sloan School of Management, M.I.T., Boston, MA, 1972, pp. 612-672,.

Peritz E. A note on multiple comparisons, Unpublished manuscript, Hebrew University, Israel, 1970.

Hsu J C. Multiple comparisons. Boca Raton : Chapman&Hall, 1996.


  • There are currently no refbacks.