pseudo R2 calculation for probit

[melike_f]

Hi all

I am using following command to estimate a panel probit model

xtprobit depvar independentvars , pa corr(ar 1) force tolerance(1e-4) vce(robust)

For the results table I would like to report beta (standardized coef and se). I used estadd listcoef with probit, but it does not run for xtprobit. So how can I list beta coef. I am not sure that estadd beta works with xtprobit

Also, xtprobit, pa option does not save scalars like pseudo r2 ll constant only or other. How can report / calculate pseudo r2 ?

thanks a lot!

[Mike Murphy]

On the first part of your question, check out the margins command with the post option specified. I believe estadd can pull the estimation results from that.

[Felix Leung]

I tried estadd beta, and it worked.

webuse union, clear

xtprobit union age grade i.south i.black i.not_smsa south##c.year, ///
    pa corr(ar 1) force tolerance(1e-4) vce(robust)

estadd beta
esttab, cells(beta(fmt(3)) se(fmt(3) par)) drop(0.*)

Why calculate pseudo R² when you can calculate R²?
Continuing from the code above:

predict union_p if e(sample), mu
correlate union union_p if e(sample)
display "R-sqaured: " r(rho)^2

You should get: .04749131.
(This is borrowed from Nick Cox's answer to this FAQ: http://www.stata.com/support/faqs/statistics/r-squared/)

But why calculate the R² when (and I quote Nick, from the same link):

There is likely to be a great deal of information about the limitations of the model, with implications for how it can be improved, in the detailed estimation results and residuals you can usually get from Stata, including graphical as well as numeric output. There is almost no such information in an R-squared.

---

To run the example above, make sure the estout package is installed (see http://repec.org/bocode/e/estout/installation.html).