Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: heckman/dataset issue

From   "Bradley, Steven W" <>
To   <>
Subject   st: heckman/dataset issue
Date   Sun, 25 Sep 2005 19:31:33 -0500

Dear Statalist:
I am trying to run a Heckman correction on a panel data set that has a significant amount of attrition in the data as there are numerous new firms that drop out during the 8 year time period.  When I run heckman, I notice that the number of unobserved to be used in heckman is much smaller (13) as shown below.

Heckman selection model                         Number of obs      =      7637
(regression model with sample selection)        Censored obs       =        13
                                                Uncensored obs     =      7624
                                                Wald chi2(6)       =    731.45
Log likelihood = -82337.21                      Prob > chi2        =    0.0000

My data set is such that when a firm drops out, there are no subsequent years entered for that firm.  Do I need to have a year for each firm even if they drop out and just enter zeros for the dependent and independent variables?
Is there a way to go through and add observations easily?
Thanks for any advice!

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index