|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: Trouble with xtivreg, fe output
Noel,
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Noel Campbell
> Sent: 30 January 2007 02:33
> To: [email protected]
> Subject: st: Trouble with xtivreg, fe output
>
> In keeping with my tradition asking foolish questions, I offer the
> following:
>
> I'm monkeying around with a variety of potential instruments.
> As is common nearly all of my candidates are strongly
> correlated with my two troublesome variables. The issue is
> this. With several specifications, using the FE estimator,
> the R-square within value is "." Just a period.
> What does that mean?
My guess is that this is the same issue that can arise with straight IV,
i.e., that the R-sq is negative. See
http://www.stata.com/support/faqs/stat/2sls.html
And btw, if you use -xtivreg2- instead of -xtivreg-, you'll see the
negative R-sq displayed, along with some rather more useful diagnostics
such as overidentification and weak identification stats. -xtivreg2- is
downloadable in the usual way from within Stata. (Dislaimer: I'm the
author of -xtivreg2-.)
Cheers,
Mark
> That the explained variance is zero?
> That doesn't seem right.
>
> Hoping y'all can help me.
>
> thanks in advance,
>
> Noel
>
>
> *
> * 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/
