Re: st: Adding R-Squared to sw command output

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 03:53 PM 5/27/2004 +0100, Rena Jenkins wrote:
> Is there a Stata programming guru out there who could help me modify the 
> sw command so that the R-squared of the model is output at each stage? (I 
> am just running a linear regression.)  i.e. to print the value of e(r2) at 
> each stage.
>
> So that when you run something like
> sw regress y x1 x2 x3, pe(0.05)
>
> Instead of
>                       begin with empty model
> p = 0.0032 <  0.0500  adding   x2
> p = 0.0153 <  0.0500  adding   x1
> p = 0.0424 <  0.0500  adding   x3
>
> You get something like:
>                       begin with empty model
> p = 0.0032 <  0.0500  adding   x2, R-squared = 0.2018
> p = 0.0153 <  0.0500  adding   x1, R-squared = 0.2523
> p = 0.0424 <  0.0500  adding   x3, R-squared = 0.2852
Rather than replacing sw.ado, I would suggest copying it as sw2.ado, and 
place it in c:\ado\personal.  Change the line that says "program define sw" 
to "program define sw2".  Then, search for the line that includes the word 
"adding".  Replace it with

di in gr " <  " %5.4f $HSW_Pe "  adding   " in ye "${HSW_`k1'}" in gr ", 
R^2 = " in ye e(r2)

This works with ols regression; sw2 would need further coding if you wanted 
to make it more general than that. I get output like

. sw2 reg income educ jobexp black, pe(.10)
                      begin with empty model
p = 0.0000 <  0.1000  adding   educ, R^2 = .65926337
p = 0.0000 <  0.1000  adding   jobexp, R^2 = .81626138
p = 0.0000 <  0.1000  adding   black, R^2 = .82641678

If you really want blow by blow accounts of what is happening in a stepwise 
regression, SPSS is better.

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/