Dear Statalisters,
This problem has been addressed before, but to my humble understanding the
discussion has not provided any straightforward solutions.
I am working with a very large (unbalanced) panel data set in which all
dependent variables are censored. The logical tools that Stata provides for
this type of analysis, -xttobit- or -xtintreg-, however do not come with a
robust option and are therefore not very reliable when the models are
heteroskedastic (or nonnormal) which is the case for my models.
One solution provided in statalist is estimating the tobit model in glamm,
which includes a robust option
(http://www.stata.com/statalist/archive/2004-02/msg00789.html) . The
rationale behind it however is somewhat outside my understanding and the
method seems tedious and extremely time consuming.
Another solution provided is to simply use bootstrapped standard errors
(http://www.stata.com/statalist/archive/2006-06/msg00279.html) . However, as
far as I understand this doesn�t really solve the problem, but only
partially improves results at best.
Ideally there would be some form of ICLAD or CLAD routine for panel data in
Stata (http://datasets.org/statalist/archive/2003-05/msg00033.html) , but as
far as I can see this is nonexistent.
Can anyone provide another solution for my problem or provide assistance in
my understanding and/or application of one of previously mentioned
solutions?
Many Thanks in advance,
Ren�
The Netherlands
*
* 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/
