My question refers to the performance of robust Tobit estimates:
Deaton (1997) states that in the presence of heteroskedasticity the standard Tobit model results in inconsistent coefficient estimates due to the scale and location dependence in the likelihood function. From my understanding this is because the standard model allows for censoring only at low values of x but not at high values. My question is whether the intreg (plus robust) command in stata (which allows for censoring at all values) overcomes this problem and gives consistent beta estimates or should one employ the CLAD estimator developed by Powell?
Thank you for your help.
*
* 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/
