Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Re: Question re Heckman


From   "R.E. De Hoyos" <[email protected]>
To   <[email protected]>
Subject   Re: st: Re: Question re Heckman
Date   Sat, 8 Jul 2006 14:32:44 -0400

Dear Adam,

I am sorry, you are completely right -predict, xb- does NOT include the Mills. This is easily verified because the expected value of the point estimator of -predict, xb- is not equal to the mean value of the dependent variable. The difference on the means is indeed the value of the Mills times lambda.

Again, sorry for the mistake.

Rafa
________________________
Rafael E. De Hoyos
Faculty of Economics
University of Cambridge
CB3 9DE, UK
www.econ.cam.ac.uk/phd/red29/

----- Original Message ----- From: "Adam Przeworski" <[email protected]>
To: <[email protected]>
Sent: Saturday, July 08, 2006 8:44 AM
Subject: Re: st: Re: Question re Heckman



Rafa,

I solved the problem. But now I think you are wrong: predict, xb
generates values identical to those calculated without including Mills.
What makes you think that predict does include lambda?

Adam

Adam Przeworski
Carroll and Milton Petrie Professor
Department of Politics
New York University
726 Broadway, Room 722
New York, NY 10003

----- Original Message -----
From: "R.E. De Hoyos" <[email protected]>
Date: Friday, July 7, 2006 6:47 pm
Subject: st: Re: Question re Heckman

Adam,

Yes they do. To eliminate them, you can either create a vector of
coefficients without the Mills' parameter:

After heckman, type
mat B = e(b)
scalar df = (e(df_m)+)
mat coef = b(1,1..df)
mat score nomills = coef


Alternatively you can compute the Mills with option -mills-,
predict xb, use
the parameter of the Mills' ratio to multiply it by the value of
the Mills
and finally subtract it from the xb prediction.

I hope this helps,
Rafa
________________________
Rafael E. De Hoyos
Faculty of Economics
University of Cambridge
CB3 9DE, UK
www.econ.cam.ac.uk/phd/red29/





----- Original Message ----- From: "Adam Przeworski" <[email protected]>
To: <[email protected]>
Sent: Friday, July 07, 2006 10:10 AM
Subject: st: Question re Heckman


> Does anyone know whether HECKMAN predict, xb included the values
of the
> Mill ratio in the predicted values? If yes, how to eliminate it?
>
> Adam Przeworski
> Carroll and Milton Petrie Professor
> Department of Politics
> New York University
> 726 Broadway, Room 722
> New York, NY 10003
> *
> * 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/
>
*
* 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/

*
*   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/

*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index