Sabrina Di Addario <[email protected]> is using -pweight-s and
-vce(jackknife)- with the seemingly unrelated bivariate probit model:
> I am using the jackknife method to compute the SE of this biprobit:
>
> biprobit (y1= x2 y1 y2 u1) (y2 = x2 z2) [pweight=weight], robust
> cluster(clusty).
>
> But, writing:
>
> biprobit (y1=x1 x2 y2 u2) (y2 = x2 z2) [pweight=weight], vce(jackknife)
> robust cluster(clusty)
>
> gives the error-message:
> (running biprobit on estimation sample)
> varlist not allowed
> an error occurred when jackknife executed biprobit
> r(101);
>
> which disappears only if I eliminate [pweight=weight]: that is, I do get
> results for:
> biprobit (y1=x1 x2 y2 u2) (y2 = x2 z2), vce(jackknife) robust
> cluster(clusty)
>
> doe sit mean that I cannot use pweights or is there a way to put them
> somewhere else in the command?
> If not, can I do this time of estimation with some other command (e.g.,
> bootstrap) supporting the cluster, robust and pweight options?
Sabrina found a parsing error in -jackknife- that is caused by the
multi-equation specification of the SU biprobit syntax.
The fix for this will be available in an upcoming ado-file update.
In the mean time, Sabrina can specify the -pweight-s inside the -vce()- option
(an undocumented feature).
biprobit (y1=x1 x2 y2 u2) (y2 = x2 z2), ///
vce(jackknife [pweight=weight]) cluster(clusty)
--Jeff
[email protected]
*
* 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/
