Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: something like a Chow test for clogit?


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: something like a Chow test for clogit?
Date   Tue, 17 Jun 2008 11:37:23 +0100 (BST)

--- Jayme Sar <jaymesar@gmail.com> wrote:
> > I want to estimate a conditional logit for a discrete choice
> > problem, and I need to make sure that the results are consistent
> > for all subsamples in the population.

--- Maarten buis <maartenbuis@yahoo.co.uk> wrote:
> what about:
> gen d = _n <= 200
> gen dXx1 = d*x1
> gen dXx2 = d*x2
> clogit y x1 x2 d dXx1 dXx2
> test d dXx1 dXx2

--- alfredo jimenez palmero wrote to me privately:
> I am writing you directly because i have never seen a question like
> this (with my own results and asking for a reference) in statalist.

Some general advise concerning the choice between writing private mails
versus mails to the statalist is discussed in the statalist FAQ under
point 5. 

> I was also with the same problem, and  following your steps i got
> this from Stata:
> 
> . test d  dXx1 dXx2 dXx3 dXx4 dXx5 dXx6 dXx7 dXx8 dXx9 dXx10
>  ( 1)  [presenci]d = 0
>  ( 2)  [presenci]dXx1 = 0
>  ( 3)  [presenci]dXx2 = 0
>  ( 4)  [presenci]dXx3 = 0
>  ( 5)  [presenci]dXx4 = 0
>  ( 6)  [presenci]dXx5 = 0
>  ( 7)  [presenci]dXx6 = 0
>  ( 8)  [presenci]dXx7 = 0
>  ( 9)  [presenci]dXx8 = 0
>  (10)  [presenci]dXx9 = 0
>  (11)  [presenci]dXx10 = 0
>            chi2( 11) =   20.23
>          Prob > chi2 =    0.0423
> 
> As all the variables got a zero, and Prob >chi2 is smaller than 0,05
> (i usually use a 95% confidece level) can i assume that the result
> are consistentfor all subsamples?  

No, what is reported under ( 1), ( 2), etc. is the null hypothesis. A
value less than .05 means that this null hypothesis is rejected, i.e.
you found that at least one of the coefficients in the subsample
differs from the effect in the rest of the sample. 

> Have you got any reference (book, article,...) for this statement?

No, I just applied what I knew about interaction effects and
statistical test to the question. Though you might find
http://www.stata.com/support/faqs/stat/chow.html and
http://www.stata.com/support/faqs/stat/chow2.html helpful.

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      __________________________________________________________
Sent from Yahoo! Mail.
A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index