In article <000f01c02504$0878b840$[EMAIL PROTECTED]>,
David A. Heiser <[EMAIL PROTECTED]> wrote:
>1. I appreciate professor deLeuw recommending A. W. F. Edward's book
>"Likelihood" (expanded version). Read it from cover to cover. Excellent
>source of ideas and analysis of Fisher's contributions.
>2. The issue is, do we follow the maximum likelihood explicitly and
>conclude that the most likely value of the distribution parameter is at
>the maximum likelihood point; or do we hedge.
Where did you get this idea? And even if it is true, why should
this be the action to take?
One does not need to take artificial examples of problems where
the maximum value of the likelihood function is infinite, and
for which rather good procedures exist. Examples of this are
the log-normal distribution with a translation parameter, and
a mixture of two arbitrary normal distributions.
Consider the actual model and consequences of actions, and use
these to decide what action to take. One can argue that there
is no need to use anything other than the likelihood FUNCTION,
but this does not state that maximizing it with respect to the
parameters is even a reasonable thing to do.
--
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, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
[EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
