Re: st: multiple regression, r squared and normality of residuals

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

--- On Wed, 23/3/11, Arti Pandey wrote:
> I ran multiple regression with in stata using two
> models;
> the first gave an R-squared of .35, No. of obs. used
> was 84, distribution of residuals was normal.
> Then I did a log transform of the dependent variable, 
> r squared went up to .65. The residuals were also
> slightly skewed to the left. No. of obs went down to
> 77. My question is how do I decide between the R
> squared and distribution of  residuals. Is such a
> high rise in R squared worth sacrificing no of
> observations?

No, nothing is ever worth sacrificing observations, they
are the only reason why our results are (somewhat) 
credible. The model is just a way of summarizing what we
have seen, it never adds credibility to our results. So
when we have to choose between data and model we should
always choose data.

If you want to try a legitimate version of the log 
transformation, you can try -glm- together with the 
-link(log)- option.

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/