[R] censReg or tobit: testing for assumptions in R?

E Hofstadler e.hofstadler at gmail.com
Mon Feb 14 16:58:44 CET 2011


Thanks for your response!

Terry Therneau wrote:
>   I assume you mean "survreg" in the survival package.  It's a shame
> that censored gaussian regression has earned a unique label (tobit) that
> makes people think it is something so very different.

(yes, what I meant was the function tobit() from the AER package, which 
however interfaces the survreg function from the survival package:
http://rgm2.lab.nig.ac.jp/RGM2/func.php?rd_id=AER:tobit)

> M>y hopefully not too silly question is this: I understand that
>> heteroskedasticity and nonnormal errors are even more serious problems
>> in a censored regression than in an ols-regression. But I'm not sure
>> how to test for these assumptions in R? Is there a way to get to the
>> residuals of censored regression models (given that corresponding
>> functions for lm, such as rstandard, are not applicable)?
> 
> Actually it can be less of a problem, since the really large outliers
> often turn into censored observations, limiting their leverage. 
> For examination, there are good ideas in the article referenced
> by ?residuals.survreg.

Many thanks for this hint, I will definitely check it out.



More information about the R-help mailing list