Re: st: Newey estimations

[Allin Cottrell <cottrell@wfu.edu>]

On Wed, 2 Aug 2006, Evelyn Colino de Cantero wrote:

> I'm running a newey-west regression with command "newey". I 
> want also to test for structural break. I'm using the optimal 
> tests suggested by Andrews and Ploberger (1994) - the 
> Mean-Wald test -
>
> as part of the test I need to compute the standard LR test -
> or a F-test for a parameter's significance (the one measuring
> a possible change in the coeff.)
>
> If I run the command "test" after "Newey" it shows a F-test 
> result, (which is actually really huge compared with the same 
> test under a regular OLS regression on the same equation)
Check out

http://www.stata.com/help.cgi?newey

and the link from there,

http://www.stata.com/help.cgi?newey+postestimation

I think the latter is what you really want, in particular the 
"lincom" and "test" items.  The "test" item in this context 
seems to imply that you're getting a Wald test based on the 
robust estimate of the covariance matrix.  But if the robust 
F-test result is "huge" compared with that associated with the 
standard OLS covariance matrix, there may be something amiss.

Do you grok Gauss?  If so, you might look at

http://www.ssc.wisc.edu/~bhansen/progs/joe_00.html

Allin Cottrell
Wake Forest University
*
*   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/