Quantcast

st: two step heckman estimation

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: two step heckman estimation

amatoallah ouchen
Good Day stata-listers,
in order to estimate the  starting points of a model , I've found out
in one article  a methodology using  the two step Heckman methodology,
which said that i'have to estimate two  tobit model one on  the
positive value,  and the second on the negative values
the model is:
                     rt=const+b*rm+µ
so i've put
generate ripo= (rt >=0)
generate rinega= (rt <=0)
heckman rt rm, select(ripo=rm) twostep
heckman rt rm, select(rinega=rm) twostep
I'need those two models to use their parameters for my manipulations
I've got

                                         Prob > chi2        =    0.0227

------------------------------------------------------------------------------
             |      Coef.      Std. Err.        z    P>|z|     [95%
Conf. Interval]
-------------+----------------------------------------------------------------
rt            |
      rm    |   3.158874   7.525672     0.42   0.675    -11.59117    17.90892
  _cons   |   -5.54708   18.46043    -0.30   0.764    -41.72886     30.6347
-------------+----------------------------------------------------------------
ripo         |
     rm      |   .4317474   .1587585     2.72   0.007     .1205866    .7429083
  _cons    |   .4160715   .0924356     4.50   0.000     .2349012    .5972419
-------------+----------------------------------------------------------------
mills       |
   lambda |   11.18048   32.78645     0.34   0.733    -53.07978    75.44074
-------------+----------------------------------------------------------------
         rho |    1.00000
       sigma |  11.180482
      lambda |  11.180482   32.78645
------------------------------------------------------------------------------


------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
rt           |
        rm |   1.771167   4.618815     0.38   0.701    -7.281545    10.82388
   _cons |   3.144992    12.2705     0.26   0.798    -20.90474    27.19473
-------------+----------------------------------------------------------------
rinega      |
          rm |  -.4810694   .1575142    -3.05   0.002    -.7897916   -.1723473
     _cons |   .1175725   .0894371     1.31   0.189    -.0577209    .2928659
-------------+----------------------------------------------------------------
mills        |
      lambda |  -5.336007   16.77838    -0.32   0.750    -38.22102      27.549
-------------+----------------------------------------------------------------
         rho |   -1.00000
       sigma |  5.3360071
      lambda | -5.3360071   16.77838
---------------------------

I'm lost among all those figures  , I'want to know which ones
represent  the parameters of the Tow models,
Sorry if my question seems too easy here, but it  is really important for me,

many thanks in advance!!

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