|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: swaic, different models achieved with forward vs. backward
|
From |
"Nick Cox" <[email protected]> |
|
To |
<[email protected]> |
|
Subject |
st: RE: swaic, different models achieved with forward vs. backward |
|
Date |
Thu, 14 Dec 2006 22:27:08 -0000 |
The authors may have a take on this, but my rules
of thumb here are
1. To note that if it makes a difference, your
model choice is fragile. The attitude should be
not to identify which model is best, but to compare models
that appear about equally good and wonder why.
2. To re-read the section on stepwise in Frank Harrell's
book on "Regression modeling strategies" whenever
I think of doing anything stepwise.
Nick
[email protected]
Michael McCulloch
> I'm comparing forward vs. backward options for swaic, after stcox.
> Different models are achieved. Is there a rule-of-thumb about
> which to use?
*
* 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/
