Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Incomplete references [was: Re: st: predicting marginal effects in a conditional logit model with fixed effects]


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Incomplete references [was: Re: st: predicting marginal effects in a conditional logit model with fixed effects]
Date   Thu, 19 Apr 2012 09:49:48 +0100

This will be wearying familiar to many, but I have to support Maarten here.

Please remember that Statalist is a discussion list, not a help line.
A help line defines an obligation to try to answer and to be
moderately deferential. A discussion list means that people can just
ignore your question. So you need to charm people into wanting to
answer your question.

Pitching your question at a very small number of people who you
presume to be totally familiar with a specific small literature and
your precise problem is sometimes difficult to avoid, but it is best
to try to reduce that by giving information that might also interest
and instruct other people.

There are actually two incomplete references here, the other to a
question in 2005. Say someone wants to look at it in the archives. Now
they have to do a search. Much better to give a URL so they can go
straight there!

Statalist is not just a way of getting answers to personal questions
in public. Making your question more interesting and informative can
help others and increase your chance of getting a good answer, or even
any answer.

Nick



On Thu, Apr 19, 2012 at 9:06 AM, Maarten Buis <maartenlbuis@gmail.com> wrote:
> On Wed, Apr 18, 2012 at 7:31 PM, Stefanie Kneer wrote:
>> I have a question with respect to the conditional logit model with
>> individual fixed efffects for a panel.
>> Is there any chance to calculate marginal effects or something similar
>> to it for such a model in stata?
>> I tried the command
>> command mfx compute, predict (pu0)
>> which however drops the assumption of having fixed effects.
>> The last entry to this question was in 2005 so i was wondering whether
>> maybe a new method has turned up to calculate the magnitude of the
>> coefficients?
>> The reason why I am asking is because according to Ai and Norton(
>> 2003) one cannot even interprete the signs of the coefficients.
>
> The answer is in Buis (2010).
>
> -- Maarten
>
> Ps. I know I am violating Statalist FAQ rules, but this is just to
> illustrate how inconsiderate it is that Stefanie is requiring us to
> track down an incomplete reference.
>
> Pps. the answer is really in Buis (2010).
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
>
> http://www.maartenbuis.nl
> --------------------------
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index