[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
"Nick Cox" <email@example.com>
Fri, 30 Nov 2007 12:17:23 -0000
The Statalist FAQ details policy on posts that are not
answered and on the practice of reposting. Please see
No one has the responsibility to answer any
particular email. If a post does not receive
a reply, perhaps the question did not appear
interesting or answerable, or for some other
reason no one wanted to reply to your question.
For example, many members are unwilling to
write long personal tutorials that may or may
not recapitulate what you already know.
Although one repost is allowed, just repeating a question
will not make it any clearer or more interesting.
In this particular case, the implication seems
to be that one model is more appropriate for your
data than another. Without any more detail
in your original posting, I don't see what else
can be said. I guess that at least some people
refrained from saying that because they thought
it was both obvious and unhelpful.
Unnati Rani Saha
I have enrolled in stata list. I am receiving a number of emails
But surprisingly it is true that several times I sent some problem to
get help from the stata lister but no response.
I cannot guess the reason. Again I am sending please if any of you can
help me I'wd grateful.
I am sending the problem again for your help.
I have a question about the HT test for the use of of FEv/sRE
What would happen if while testing for the more efficient estimator, I
specified Re as robust? Therefore instead of
xtreg depvar indepvar 1........................indepvarn, re
xtreg depvar indepvar 1........................indepvarn, re robust?
I did this just as a test out of curiosity (without considering the full
implications of this) and I find that the differenc in the two commands
changes the Prob>chi2 dramnatically so that in the former case FE is
appropriate and in the latter RE is appropriate
* For searches and help try: