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]

Re: st: Heteroskedastic Probit Model


From   Mustafa Brahim <datotanseri@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Heteroskedastic Probit Model
Date   Thu, 22 Apr 2010 19:57:51 +0800

Thank you Maarten ! Let me clarify few points.

you mentioned that "most of the information is comming
from your assumptions (primarily functional form assumptions)" and you
conclude that I do not have a good theoretical background. Well let me
tell you that in all studies related to my topic no one mentioned or
tested the model for heteroskedasticity and the functional form I am
using is similar to what is well known in the field. And I remember
you telling me once that we cannot take something for granted simply
because many researchers did it. So because no one tested the model
for heteroskedasticity does not mean thatI should not test mine.
you suggested not to use hetprob,the question then is  how am I going
to check whether it exists or not? or should I just ignore it and
assume homoskedasticity? to be honnest I am not sure what to do? so
far I got to know that the only way to check for heteroskadasticity in
probit model is to use hetprob. If I don't use hetprob what should I
do then to test for robustness of the model? if someone can clarify
this that would be very helpful.

Regards,

Ismail

On Thu, Apr 22, 2010 at 5:54 PM, Maarten buis <maartenbuis@yahoo.co.uk> wrote:
> --- On Thu, 22/4/10, Mustafa Brahim  wrote:
>> 1) I understand that runing hetprob procedure adjusts the
>> estimates and standard errors and therefore corrects for
>> heteroskedasticity? is this correct?
>
> That is true only in a very specific sense. Remember that
> the dependent variable is latent, so it is not directly
> observed. The errorterm is the difference between a
> predicted value and a latent variable. The
> heteroskedasticity means that the variance of this error
> term changes over some of the variables. Notice that our
> data contains extremely little information on this
> heteroskedasticity, so most of the information is comming
> from your assumptions (primarily functional form
> assumptions).
>
> So, when should you use -hetprob-? When you have a very
> strong theoretical argument for these assumptions. Given
> that you started your question with the statement: " I
> included all variables in the -het( )- option because I am
> not sure which one may be causing the problem of
> heteroskedasticity." I conclude that you do not have such
> a theory, and should thus not use -hetprob-.
>
> Hope this helps,
> Maarten
>
> --------------------------
> 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