Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Outlier diagnostics for tobit (postestimation)


From   "Timo Beck" <timoworldwide@gmx.de>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Outlier diagnostics for tobit (postestimation)
Date   Fri, 19 Oct 2012 13:51:44 +0200

Dear Martin,

Great! Thank you! That helps a lot! One last question regarding this topic. Just looked at the code and wondered about one little thing. As I understood, dfbetas are calculated simply by dividing the difference between the two coefficient estimates with and without the observation by the standard deviation. Why would one inlcude another division by e(N)? 

@Nick: Fair enough! Thanks!

Timo


-------- Original-Nachricht --------
> Datum: Fri, 19 Oct 2012 11:54:01 +0200
> Von: Maarten Buis <maartenlbuis@gmail.com>
> An: statalist@hsphsun2.harvard.edu
> Betreff: Re: st: Outlier diagnostics for tobit (postestimation)

> On Fri, Oct 19, 2012 at 11:48 AM, Timo Beck wrote:
> > Once again quickly re my other question, maybe you also have an opinion
> on whether, just as a robustness test, I could fit OLS as an approximation
> of the tobit model and use outlier diagnostics thereafter and then simulate
> the tobit without these identified cases? Or would I be doing something
> completely wrong? According to Wooldridge a linear model is a good
> approximation for E(y) in a corner solution model which is what I am looking at.
> That's why I am thinking that way.
> 
> I would not do that. It is possible that a linear regression is a good
> approximation of a tobit model in your data, but there is no
> guarantee. Moreover, it is not necessary as you can compute dfbetas in
> a tobit model, as I have shown before:
> <http://www.stata.com/statalist/archive/2012-10/msg00865.html>
> 
> -- Maarten
> 
> ---------------------------------
> Maarten L. Buis
> WZB
> Reichpietschufer 50
> 10785 Berlin
> Germany
> 
> http://www.maartenbuis.nl
> ---------------------------------
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index