st: Need help with Heckprob (rho)

[Thomas Schmeling <tschmeling@ric.edu>]

Hello,

I'm new to stata, new to this list, and at the limit of my training with regard to these techniques, but hope someone can lend a hand.

I've run Heckprob and am having trouble interpreting the results, which I've pasted in below. The coefficients all look right, but RHO has me terribly puzzled.  Can it actually be 1, or is this a symptom of a problem?  Given the std. err. of Rho, how can the confidence interval be -1 to 1?  I'll note that I had to use the "difficult" option, as this model generated a lot of "not concave" messages.  Also, I have no idea how to interpret /athrho, and can't find a word of explanation in the documentation. Any pointers?
Thanks,
Tom Schmeling



Probit model with sample selection              Number of obs      =      1592
Censored obs       =    1537
Uncensored obs     =    55

Wald chi2(3)       =    220.47
Log likelihood = -255.6258                      Prob > chi2        =    0.0000

------------------------------------------------------------------------------
|      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
ctallow      |
prenat |   .3497772    .076645     4.56   0.000     .1995557    .4999987
negprec |  -.3912559   .1384237    -2.83   0.005    -.6625613   -.1199505
cuma |   .0151733   .0070801     2.14   0.032     .0012966      .02905
_cons |   -2.25036    .132718   -16.96   0.000    -2.510482   -1.990237
-------------+----------------------------------------------------------------
anycase      |
negprec |  -.4459885   .1281523    -3.48   0.000    -.6971624   -.1948147
yrsofpre |  -.0177743   .0094776    -1.88   0.061      -.03635    .0008014
cuma |   .0438021   .0236192     1.85   0.064    -.0024907    .0900948
logpop |   .5996745   .1005904     5.96   0.000      .402521    .7968281
year1 |  -.0148024   .0169564    -0.87   0.383    -.0480364    .0184316
_cons |  -5.053822   1.206477    -4.19   0.000    -7.418472   -2.689171
-------------+----------------------------------------------------------------
/athrho |   11.62408   224.1089     0.05   0.959    -427.6213    450.8695
-------------+----------------------------------------------------------------
rho |          1   7.18e-08                            -1      1
------------------------------------------------------------------------------
LR test of indep. eqns. (rho = 0):   chi2(1) =    11.16   Prob > chi2 = 0.0008
------------------------------------------------------------------------------
 --------------------------
Dept. of Political Science
Rhode Island College  
600 Mount Pleasant Ave. Providence, RI 02908   
Phone: (401) 456-8722
* *   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/