Wrong, as the email from Richard Williams
indicates.
Nick
[email protected]
Nick Cox
> I doubt you're missing anything pre-existing.
>
> One reason for that, I guess, is that this
> puts a great deal of emphasis on one questionable
> figure of merit.
>
> Another reason is that you can knit your own.
>
> Here's one template.
>
> sysuse auto, clear
> tempname r2
> scalar `r2' = 0
>
> local blocks `" "headroom weight" "turn gear price" "rep78 trunk" "'
> local blocks `" `blocks' "length displacement foreign" "'
>
> foreach x of local blocks {
> local X "`X' `x'"
> qui regress mpg `X'
> di as txt "`x'{col 30}" _c
> di as txt "R-sq " as res %6.4f e(r2) _c
> di as txt " Change " as res %6.4f e(r2) - `r2'
> scalar `r2' = e(r2)
> }
>
> headroom weight R-sq 0.6523 Change 0.6523
> turn gear price R-sq 0.6599 Change 0.0077
> rep78 trunk R-sq 0.6690 Change 0.0091
> length displacement foreign R-sq 0.7155 Change 0.0466
>
> Nick
> [email protected]
>
> Garrard, Wendy M.
>
> > I need to do a regression using blocks of predictors, and report the
> > change in R^2 with the addition of each block. I can't identify a
> > regression command for doing this using the obvious keyword
> > searches of
> > help files or the manual. Is there a standard command that I am
> > overlooking? Or is there a user-written program for this?
>
> *
> * 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/
>
*
* 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/
