Thanks, Clive. I take it that bw(2) and not bw(1) must be used to
model an AR1 process. The only difference I see between ivreg and
newey is that ivreg makes the "small" adjustment for the degrees of
freedom. But neither approach addresses the issue of contemporaneously
correlated errors, right?
Cheers,
Mark
On Sun, Mar 2, 2008 at 1:26 PM, Clive Nicholas
<[email protected]> wrote:
> Mark Dincecco wrote:
>
> > I have an unbalanced panel data set with a very large T to N ratio.
> > Essentially, I follow 5 countries over a period of 200 or more years.
> > I must correct for contemporaneously correlated errors, panel
> > heteroskedasticity, and common serial correlation. I have opted to use
> > newey2 or xtpcse rather than xtgls following Beck and Katz (1995).
> >
> > My question is the following: Is the only difference between newey2
> > and xtpcse that xtpcse controls for contemporaneously correlated
> > errors while newey2 does not? That is, does newey2 control for
> > contemporaneously correlated errors? Please note that for both
> > techniques I already control for panel heteroskedasticity and common
> > serial correlation.
>
> [...]
>
> I think your reading of the differences are broadly correct, but
> instead of using -newey2-, why not use -ivreg2-, downloadable from
> SSC, instead? Then you could do:
>
> . ivreg2 y x1 x2, bw(2) robust small
>
> which gives you OLS estimates with Newey West standard errors.
>
> --
> Clive Nicholas
>
> [Please DO NOT mail me personally here, but at
> <[email protected]>. Thanks!]
>
> "Courage is going from failure to failure without losing enthusiasm."
> -- Winston Churchill
>
>
> *
> * 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/
