Re: st: Re: stata question

[Cordula Stolberg <cs32@sussex.ac.uk>]

Rose,

If you just want to do a regression without instrumental variables, you 
might also try to use newey2 (you can install it after typing -findit 
newey2-) with the -force- option, i.e.

newey2 y x1 x2 x3, lag(#) force

If you have an fe model, you will have to add the dummies yourself.

Hope this helps.
Cordula

--On 17 August 2004 12:04 -0400 Kit Baum <baum@bc.edu> wrote:


> Rose,
>
> Suggest you use ivreg2 (findit ivreg2). It allows for a variety of HAC
> error treatments (one of which is the Bartlett kernel used by
> Newey-West), and works with panel data. If you want dummies (as in a FE
> model), you will have to provide your own (or do the demeaning
> previously). Please note that for HAC you must specify both robust and
> bw() options.
>
> Best wishes
> Kit Baum
>
> On Aug 17, 2004, at 11:57 AM, Rose Zhao wrote:
>
> > Hi Chris,
> >
> > I have a stata question and  I am wondering if you can help---
> >
> > how can you run a regression on a time-serie cross-sectional dataset
> > (panel dataset) using Newey robust errors in stata? I can't get the
> > newey command to work.
> >
> > Thanks,
> >
> > --
> > Rose
> *
> *   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/



*
*   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/