Statalist


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

Re: st: regression using eform


From   Clive Nicholas <clivelists@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: regression using eform
Date   Thu, 3 Sep 2009 20:54:21 +0100

Tina Hernandez-Boussard wrote:

> I have a question regarding a regression model that I am running.  I am
> looking at predictors for a variable slope, which is growth of a tumor per
> day.  I have ran this model and using
>
> sw, pr(.2):regress slope smoker exercise bmi , eform(odds)
>
> I have reran the model, only multiplying the slope by 365.25 to get the
> growth per year.  I did not think that this would change anything, yet it
> changes the odds ratios.  I still get the same prob>f, r-squared, and
> adjusted r-squared.  However, model and residual ss, df, and ms are
> different.
>
> Can someone please explain why my odds ratios are changing, yet the t and
> P>|t| are not?

[...]

I have two brief recommendations to make:

(1) Stepwise regression is generally disparaged. Why? Look at:
http://www.stata.com/support/faqs/stat/stepwise.html

(2) Look at http://www.imperial.ac.uk/nhli/r.newson/papers/gmratio.pdf
if you want to gain a quick understanding of how to interpret results
using -regress- with the -eform()- option.

-- 
Clive Nicholas

[Please DO NOT mail me personally here, but at
<clivenicholas@hotmail.com>. Please respond to contributions I make in
a list thread here. Thanks!]

"My colleagues in the social sciences talk a great deal about
methodology. I prefer to call it style." -- Freeman J. Dyson.

*
*   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   |   What's new   |   Site index