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
<[email protected]>. 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/
