# st: Re: Need help with Heckprob (rho)

 From "Scott Merryman" <[email protected]> To <[email protected]> Subject st: Re: Need help with Heckprob (rho) Date Thu, 24 Jul 2003 20:55:20 -0500

```----- Original Message -----
From: Thomas Schmeling
To: [email protected]
Sent: Thursday, July 24, 2003 5:16 PM
Subject: st: Need help with Heckprob (rho)

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

Tom,

It is likely the symptom of a problem.  This very problem is discussed in the
manual [R]heckman, page 71.

Rho is bound between -1 and 1;  it is the correlation between the regression
equation and the selection equation.

Also, typing -findit heckman- brings up a number very good FAQ dealing various
issues of the heckman model:

FAQ     . . . . . . . . . . . . . . . . . .  Define constraints for parameters
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  W. Guan
11/01   How do I impose the restriction that rho is zero
using the heckman command with full ml?
http://www.stata.com/support/faqs/stat/constraints.html

FAQ     . . . . . . . . . . . . . . . Endogeneity versus sample selection bias
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .  D. Millimet
10/01   What is the difference between 'endogeneity' and
'sample selection bias'?
http://www.stata.com/support/faqs/stat/bias.html

FAQ     . . . . . . . . . . . . . . Determining the sample for a Heckman model
. . . . . . . . . . . . . . . . . . . . . . .  V. Wiggins and W. Gould
3/01    Why are observations that are noninformative about
the dependent variable, but are known to be selected,
excluded by heckman from the estimation sample?
http://www.stata.com/support/faqs/stat/heckman.html

FAQ     . Computation of rho in the two-step Heckman estimator. Technical FAQ.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Wiggins
3/00    How are estimates of rho outside the bounds [-1,1]
handled in the two-step Heckman estimator?
http://www.stata.com/support/faqs/stat/twosteprho.html

FAQ     . Mills' ratios and censoring direction in the Heckman selection model
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Wiggins
5/99    Why are there so many formulas for the inverse of
Mills' ratio?
What if I have censoring from above/below in my
Heckman selection model?
http://www.stata.com/support/faqs/stat/invmills.html

Hope this helps,
Scott

*
*   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/
```