Significance Difference Test Between Two Treatments with Changing Population Mean Vectors

Accession number;03A0275216
Title;Significance Difference Test Between Two Treatments with Changing Population Mean Vectors
Author; NAKAMURA T (Tokai Univ., Kanagawa, Jpn) MOTOYAMA K (Tokai Univ., Kanagawa, Jpn) DOUKE H (Tokai Univ., Kanagawa, Jpn)
Journal Title;Proc Sch Sci Tokai Univ
Journal Code:S0439A
ISSN:0919-5025
VOL.38;NO.;PAGE.71-79(2003)
Figure&Table&Reference;TBL.2, REF.5
Pub. Country;Japan
Language;English
Abstract;The study is to propose a procedure to test the significance difference between two treatments where the population mean vectors change with levels by using the concept of the profile analysis. We suppose that the response vector on each level for two treatments is distributed according to the multivariate normal distribution with known covariance matrix. We realize the procedure by using the test statistic based on the design matrices. The procedure is possible to carry out our profile analysis between two treatments without using the asymptotic distribution. (author abst.)
FULLTEXT