Postestimation saving results for ps matching

[Taylor Walter]

Hello, for teffects psmatch (or psmatch2, if easier), is it possible to use have Stata save the main coefficients and p-values so that I can use regsave? I tried post: with no success.
Would be something like:

teffects psmatch, post: (y) (x1 x2)
regsave using 'example', pval detail(all)

Edit: I see in the teffects psmatch helpfile that r(table) is saved as a matrix containing coefficients and p-values, perhaps easiest to call that? Not sure how to in regsave, though, as I think regsave may only be able to collect e().

[Øyvind Snilsberg]

teffects psmatch stores the coefficient vector from the propensity score model in e(bps). the option coefmat(matname) instructs regsave (SSC) to retrieve coefficient estimates from matname instead of e(b).

. webuse cattaneo2, clear
(Excerpt from Cattaneo (2010) Journal of Econometrics 155: 138–154)

. teffects psmatch (bweight) (mbsmoke mmarried c.mage##c.mage fbaby medu)

Treatment-effects estimation                   Number of obs      =      4,642
Estimator      : propensity-score matching     Matches: requested =          1
Outcome model  : matching                                     min =          1
Treatment model: logit                                        max =         74
----------------------------------------------------------------------------------------
                       |              AI robust
               bweight | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------------+----------------------------------------------------------------
ATE                    |
               mbsmoke |
(Smoker vs Nonsmoker)  |  -210.9683     32.021    -6.59   0.000    -273.7284   -148.2083
----------------------------------------------------------------------------------------

. regsave using results, coefmat(e(bps)) replace
file results.dta saved

. use results, clear

. list, clean

                         var        coef   stderr      N  
  1.        mbsmoke:mmarried   -1.145706   32.021   4642  
  2.            mbsmoke:mage     .321518        .   4642  
  3.   mbsmoke:c.mage#c.mage   -.0060368        .   4642  
  4.           mbsmoke:fbaby   -.3864258        .   4642  
  5.            mbsmoke:medu   -.1420833        .   4642  
  6.           mbsmoke:_cons   -2.950915        .   4642

[Taylor Walter]

Thanks Oyvind Snilsberg. Two quick questions:

First, how can I save the average treatment effect (so in your example, -210.9683)? That is the value that I am hoping to save with regsave.
Second, how does it store the pvalues? Am I able to get those for the average treatment effect as well?

[Øyvind Snilsberg]

oh, I see. just run regsave without coefmat option. add pval option to get p-values.

. webuse cattaneo2, clear
(Excerpt from Cattaneo (2010) Journal of Econometrics 155: 138–154)

. teffects psmatch (bweight) (mbsmoke mmarried c.mage##c.mage fbaby medu)

Treatment-effects estimation                   Number of obs      =      4,642
Estimator      : propensity-score matching     Matches: requested =          1
Outcome model  : matching                                     min =          1
Treatment model: logit                                        max =         74
----------------------------------------------------------------------------------------
                       |              AI robust
               bweight | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------------+----------------------------------------------------------------
ATE                    |
               mbsmoke |
(Smoker vs Nonsmoker)  |  -210.9683     32.021    -6.59   0.000    -273.7284   -148.2083
----------------------------------------------------------------------------------------

. regsave using results, pval replace
file results.dta saved

. use results, clear

. list, clean

                     var        coef   stderr       pval      N  
  1.   ATE:r1vs0.mbsmoke   -210.9683   32.021   4.44e-11   4642

help regsave

[Taylor Walter]

Thank you very much!