RE: st: non-normal residual

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- Fabio Zona wrote
> 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...)

The difference between the postions has more to do with terminology than
substance. In a graphical sense --- looking for a bell shape --- it is 
the residuals that have to be normal. In a modeling sense we can very
well think of a normal distributed y where the mean changes over some
explanatory variables. Problem is, if we graph such a distribution it 
will look nothing like the bell curve that we are used to. So for 
model checking purposes the graphical sense, i.e. looking at the 
residuals rather than the dependent variable, is the "right" answer.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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/