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 at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Robust regression - outliers


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Robust regression - outliers
Date   Thu, 22 Sep 2011 19:00:09 +0100

This presupposes that outliers are easily defined, which isn't always
true. For example, it is not even always true that an outlier is
associated with a large residual, but as you mention robust regression
let's start optimistic.

In my view the best ways to look for outliers are graphical.

. search modeldiag

and read the 2004 article and install the updated software for some
suggestions, The article is directly accessible at

http://www.stata-journal.com/sjpdf.html?articlenum=gr0009

For example, -rvfplot2- in this package works after more
regression-type commands than the official -rvfplot-.

-extremes- is from SSC and not especially helpful here.

Please note that the request to explain where user-written software
you refer to come from applies to you too. (See also your last post.)

Nick

On Thu, Sep 22, 2011 at 6:46 PM, Ozgur Ozdemir <ozdemirozgur@hotmail.com> wrote:

> I am trying to ding outliers in my regression analysis but none of the proposed commands such as cooksd, rstudent, hat etc works after a robust regression. Is there any specific command to run to find the outliers? I have also seen the extremes command but not sure if i can use. thanks

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


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