Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: heckprob dropping observations


From   Eric Lawrence <eric.d.lawrence@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: heckprob dropping observations
Date   Sun, 14 Feb 2010 10:25:10 -0500

Carli,

Your estimated rho, -.999, with a standard error of .39 but a CI of -1 to 1
is telling you that your data are not informative enough to identify the
cross-equation correlation.

I highly recommend  tracking down Butler's "Estimating the Correlation in
Censored Probit Models," The Review of Economics and Statistics, Vol. 78,
No. 2 (May, 1996), pp. 356-358, which has some suggested actions in such
cases.

Eric Lawrence

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index