Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: DFBETA for constant


From   Weihua Guan <wguan@stata.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: DFBETA for constant
Date   Fri, 09 Aug 2002 14:42:56 -0500

--Roger Harbord <Roger.Harbord@bristol.ac.uk> asked early this week why
-dfbeta or -predict, dfbeta()- does not work for the constant term.

After exchanging emails with Roger, I privately sent him a do-file to
calculate the DFBETAs for the constant term based on a draft by Bobby
Gutierrez.  Now I converted the do-file to an revised version of -dfbeta-,
called -dfbeta2-.  If anyone is interested, it can be downloaded by:

. net from http://www.stata.com/users/wguan/


The syntax is basically the same as -dfbeta-, where you can use:

. dfbeta2 [varlist]

to obtain the DFBETAs for the variables specified.  I also added two options:

1) -double- requests that the calculated data will be stored as double
precision.  The default is float.

2) -constant- requests the DFBETAs for the constant term to be calculated.

Some examples:

. dfbeta2, constant		/* for the constant term only */

. dfbeta2 weight, constant	/* for weight and constant term */

. dfbeta2			/* for every variable in previous model
				   including the constant term */

Weihua Guan <wguan@stata.com>
Stata Corp.
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index