st: Re: rolling chow test

[Kit Baum <baum@bc.edu>]

What you need to do, within your 'wrapper' program, is to save both  
the sum of the coefficients and their standard error (or the p-value  
of the test statistic from lincom) in the -rolling- sample. Of course  
you cannot use -test- after running -rolling-: -rolling- creates a  
new dataset, each observation of which corresponds to one of the  
windows. What would you test? You should be generating a test  
statistic for each period (that is, the sum of the coefficients, the  
standard error of that quantity, their ratio and the p-value of that  
ratio vs an appropriate t distribution).

Happy Thanksgiving-
Kit

Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On Nov 21, 2007, at 2:33 AM, statalist-digest wrote:


> Thank you, Kit for your answer.
>
> I succeeded in getting the sum of the coefficients. But if I want  
> to use any
> command as e.g. "test" I receive error r(301), last estimates not  
> found.
> What can I do?
>
> Thanks again,
> Nadine
>
>
>
> - -----Ursprungliche Nachricht-----
> Von: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu]Im Auftrag von Kit Baum
> Gesendet: Dienstag, 20. November 2007 00:50
> An: statalist@hsphsun2.harvard.edu
> Betreff: st: re: rolling chow test
>
>
> Nadine wants to know how to do a rolling chow test. My response to
> Bethany in July covered this topic:
>
> ttp://www.hsph.harvard.edu/cgi-bin/lwgate/STATALIST/archives/
> statalist.0707/date/article-556.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/