Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: Outlier diagnostics for tobit (postestimation)

From   Maarten Buis <>
Subject   Re: st: Outlier diagnostics for tobit (postestimation)
Date   Fri, 19 Oct 2012 13:57:26 +0200

On Fri, Oct 19, 2012 at 1:51 PM, Timo Beck wrote:
> 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)?

The computation is explained here:

-- Maarten

Maarten L. Buis
Reichpietschufer 50
10785 Berlin

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index