In article <[EMAIL PROTECTED]>, Peter Hai <[EMAIL PROTECTED]> wrote: >m and n observations were obtained, respectively, from two normal >distributions which are independent of each other. How can I test the >null hypothesis that the absolute mean difference [abs(mu1-mu2)] + >variance ratio [Variance1/Variance2] is less than or equal to a >constant.
Not easily. For one thing, the mean difference has as units the units of the observations, and the variance ratio is a pure number. Also, you have a composite hypothesis; even the boundary is composite. If m and n are large, one could do a likelihood ratio test, using the MLE on the boundary and the MLE. But is what you have stated your actual hypothesis? -- This address is for information only. I do not claim that these views are those of the Statistics Department or of Purdue University. Herman Rubin, Department of Statistics, Purdue University [EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558 . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
