[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: Est. of variance for non-linear comb. of parameters (Krinsky-Robb)
Arne Risa Hole <firstname.lastname@example.org> wrote:
> Hi Jesper
> You can calculate the bootstrap variances (and CIs) of your WTP
> measures by runnning the following command:
> bootstrap (-_b[H1]/_b[H5]) (-_b[H2]/_b[H5]) (-_b[H3_varm]/_b[H5]) ///
> (-_b[H3_kold]/_b[H5]) (-_b[H4]/_b[H5]) (-_b[_cons]/_b[H5]), ///
> reps(1000): xtlogit choice H1 H2 H3_varm H3_kold H4 H5, re
> I have written a Stata module that calculates Krinsky Robb CIs for WTP
> measures which should be straightforward to tweak into calculating the
> variance as well. I will send this to you off-list.
> Best wishes
Arne had exactly what I was looking for, thanks a lot!
-- J. K. Hansen
* For searches and help try: