p s upadrashta wrote:
>
> Although apparently "a mixture of two normals, differing in means, can still
> be unimodal" (orig. Everitt, 1981 "Bimodality and the nature of
> depression") -- which I don't quite 'see' yet.
An interesing exercise in first-year calculus/second-year math stats: A
mixture of two normal distributions, each with standard deviation 1, and
withmeans differing by a, has PDF
1/sqrt{2 pi} {1/2 exp{-x^2/2} + 1/2 exp(-(x-a)^2/2} .
For what values of a is this unimodal? And sketch the limiting case.
-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/ .
=================================================================
