Re: st: ivreg2

[Austin Nichols <austinnichols@gmail.com>]

David Parsley <David.Parsley@owen.vanderbilt.edu>:
Look at the first-stage residuals with:
reg x5  x1 x2 x3 x4 i.year z1 z2 z3 z4
predict x5hat
reg x6  x1 x2 x3 x4 i.year z1 z2 z3 z4
predict x6hat
scatter x5hat x6hat

On Sat, Mar 23, 2013 at 9:34 PM, David Parsley
<David.Parsley@owen.vanderbilt.edu> wrote:
> After issuing an Ivreg2 command I try to access the post-estimation dialogs "Statistics>Postestimation>Prediction, Residuals, etc."  but I get the error: "Unable to invoke predict dialog (error 601)"
>
> Specifically I am using Stata 11.2, & I have panel data with > 1000 panel ids, and up to 44 time series observations (unbalanced). My command looks like:
> xi: ivreg2 y  x1 x2 x3 x4 i.year  (x5 x6 = z1 z2 z3 z4), first ffirst savefirst robust bw(12) endog(x5 x6)
>
> Any thoughts on what may be preventing me from getting that dialog? Since I don't know the commands available therein, I can't replicate them on the command line. I was looking for some way to look at the residuals of the first stage - maybe some scatter plots of some sort. Thanks much.
>
>
> David Parsley
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/