-nestreg- (was: st: saving a sample after a regression)

["Michael S. Hanson" <mshanson@mac.com>]

On Mar 12, 2007, at 3:56 PM, Richard Williams wrote:

> Consider using nestreg, e.g.
>
> nestreg: reg y x z
>
> This will first regress y on x and then y on x and z.  It will use 
> listwise deletion, and for an added bonus it will give you incremental 
> f tests at the end testing the vars added at each stage.
	So I am very intrigued by -nestreg- as mentioned by Richard:  I am 
working on an application right now that could benefit from this tool.  
Looking in the on-line help, it does not support the estimation 
commands I am interested in using (newey or ivreg), but I notice that 
it is possible to write one's own programs to run with -nestreg-.  
Great!  Unfortunately, the documentation for doing so does not appear 
to exist;  my copies of the version 9 [R] manuals has no entry for 
-nestreg- -- as indicated in the on-line help file, which says 
"forthcoming".  The on-line help also makes reference to using -sw- or 
-swml- as -program- properties for the user-written program.... But 
neither the on-line help nor the manuals have turned up any information 
on such properties (although I concede I may have missed them).

	So, can anyone recommend further information about implementing 
user-written commands with -nestreg-?  Thanks!

                                        -- Mike

*
*   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/