Peter Hai 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.
Why would you test, or even compute such a thing? The mean difference
and variance ratio are dimensionally different and cannot meaningfully
be added. Throw in for good measure the fact that the "null"
hypothesis is not a point hypothesis so that the relevant probabilities
cannot be computed, and there's not much left over...
-Robert Dawson
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
