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

Re: st: Censored panel data models with heteroskedasticity

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: Censored panel data models with heteroskedasticity
Date   Thu, 24 Jan 2008 11:57:07 +0000 (GMT)

--- Ren� Wevers <[email protected]> wrote:
> This problem has been addressed before, but to my humble
> understanding the discussion has not provided any straightforward
> solutions. 

The reason is that this is a non-trivial problem. Basically, you want
Stata to do the right thing even if your model is wrong. You cannot
hope that any software will give you a straightforward solution to such
a problem. For as far as I know you have listed the available options
open to you. I have just one more comment in response to the following
> 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.

The name of the -robust- option sometimes leads to too much confidence
in the results, even though the limitations are well documented in the
User's guide: [U] 20.15. It is no miracle cure, the results are robust
in only a very limited sense. All it does is change the standard
errors, and change the interpretation of your results to purely
descriptive results (though it leaves the numerical values of the
estimates unchanged). 

Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
*   For searches and help try:

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