Keep in mind that the tobit model will produce inconsistent estimates
if there is heteroskedasticity. You might want to take a look at
CLAD, censored quantile regression, or symmetrically censored least
squares.
For CLAD : -findit clad- to locate and download.
CQR: versions can be found at:
http://faculty.chicagogsb.edu/timothy.conley/research/qrcode/quantile.html
or http://www.stata.com/statalist/archive/2006-09/msg00379.html
SCLS: http://elsa.berkeley.edu/~kenchay/
Scott
On Jan 30, 2008 11:08 AM, Anita <[email protected]> wrote:
> Dear Statalistmembers
>
> I currently am doing research on testing the tobit model on
> heteroscedasticity.
> In previous postings I learned that I can get robust estimates using the
> command "intreg". (is explained in the FAQ)
> As I want to test if heteroscedasticity is present I want to do an
> LR-test, as in the description of intreg it is mentioned that it is a
> generalization of the tobit command.
> So I thought I could use it as the unrestricted model.
> I am not sure if this is the right thing to do and if it is, I would
> have to explain, what inreg actually does. (I don't know exactly, why it
> is a generalization).
> Could someone give an answer to this question? I hope it is not to silly
>
> Thanks, Anita
*
* 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/
