Jay Warner wrote:
> When I am given the estimated sigma from somewhere else, so that I can't assess
> the validity of it, I call this an _external_ estimate of sigma. I accept it with
> the usual reservations, and proceed to do the problem. The equation will be a z
> test type.
I don't think "not being able to assess the validity" of an external
estimate is enough justification to ignore the evidence about sigma
provided by the data. You are only justified in doing that, I think,
when you *are* able to assess the external estimate as being highly
reliable.
-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/ .
=================================================================
