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

st: Re: rolling chow test

From   Kit Baum <>
Subject   st: Re: rolling chow test
Date   Thu, 22 Nov 2007 09:42:35 -0500

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 Baum, Boston College Economics and DIW Berlin
An Introduction to Modern Econometrics Using Stata:

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,

- -----Ursprungliche Nachricht-----
[]Im Auftrag von Kit Baum
Gesendet: Dienstag, 20. November 2007 00:50
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:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index