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]

st: RE: non-normal residual


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: non-normal residual
Date   Thu, 29 Apr 2010 18:40:07 +0100

At most, it may be assumed that (*) errors are distributed normally. However, this is the least important assumption for most statistical purposes. 

No flavour of a standard linear model assumes that the response is distributed normally. 
It's pretty much diagnostic of lousy textbooks or articles if they say this. 

If (*) is true, then it is to expected that residuals will be distributed approximately normally. 

This much should be conveyed, and more precisely, by any decent text. 

It is not easy to give good advice on what you should do. It's not out of the question that your model is fine, or as good as could be done. But I would spend much more time looking at the residual vs fitted plot than at the distribution of residuals. 

Nick 
n.j.cox@durham.ac.uk 

Fabio Zona

I have a cross-section multiple regression: however,
- Y is not distributed normally (some say it should be; some say it is not needed! where is the truth? I read residuals must be normally distributed, not the Y...)
- any tranformation of Y does not allow to approach a normal-distribution
- beyond Y, residuals are non-normally distributed

Any suggestions on how to handle this situation?

*
*   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