Re: Selection of explanatory variables in a model
- From: "Reef Fish" <large_nassua_grouper@xxxxxxxxx>
- Date: 28 Nov 2006 03:36:14 -0800
Cat wrote:
Dear colisters,
- I heard / read (on the Internet) that the number of variables to put
as explanatory characters must not be greater than n/10 or n/ 20 (n is
the sample size). Does someone can provide me with a serious
bibliographic reference about this ?
That's someone's rule of thumb -- not an unreasonable one, but
surely not something worthy of a serious bibliographical reference.
Besides, does it really matter whose rule of thumb it was that
you heard/read (on the internet)? :-)
The other questions of yours are equally unanswerable in the
sense of anything other than anyone's opinion, without any
well-defined meaning or criterion. That is my opinion. Any
other response is equally valid and useless. IMHO.
-- Reef Fish Bob.
- Second, I've got a first set of about 15 variables (for around 150
patients), interactions excluded. I categorized some into binary
characters to yield odds-ratios. What is the risk ? lack of power ?
- Third, performing a second selection via the stepwise algorithm, is
there a consensus about the significance cut-off (alpha). to use ? I
read 20% instead of 5%. Is this usual ?
- Regarding SAS programming (I run the version 8.2 under Windows OS),
which procedure, between Proc Logistic and Proc GLM, should I choose,
and according to which criteria ?
Thanks a lot.
Catherine.
.
- Follow-Ups:
- Re: Selection of explanatory variables in a model
- From: Bruce Weaver
- Re: Selection of explanatory variables in a model
- References:
- Prev by Date: Re: How many chess games are there?...and who wins?
- Next by Date: bootstrapping
- Previous by thread: Selection of explanatory variables in a model
- Next by thread: Re: Selection of explanatory variables in a model
- Index(es):
Relevant Pages
|
Loading