Re: Selection of explanatory variables in a model




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.

.



Relevant Pages

  • Re: The Internet.
    ... Pt stated opinion, and my opinion agrees. ... This doesn't make the Internet ... some quality control and peer review involved, ... doesn't make anyone any smarter. ...
    (alt.guitar.beginner)
  • Re: ITV and its future
    ... this posting would be no reason to fire someone because everyone's ... I'm entitled to my opinion yes, ... A simple nice request to your *GATEWAY* provider (as there ... that when you used an internet gateway your basic details were publically ...
    (uk.tech.broadcast)
  • Re: "Fraud" at Dearborn
    ... > "You're a smartass on the internet ... > with an undereducated opinion, ... > "What kind of a lawyer has ZERO ...
    (sci.med.diseases.lyme)
  • Re: LS-4 Water bag repair
    ... -Whether or not this internet capability is a "primary benefit" is ... -Intimating that someone should perform their own independent research ... rather than asking a question of a web-based forum is an opinion ...
    (rec.aviation.soaring)
  • Re: LS-4 Water bag repair
    ... -Whether or not this internet capability is a "primary benefit" is ... rather than asking a question of a web-based forum is an opinion ... wholly dissimilar to qualifying the capability of internet technology. ...
    (rec.aviation.soaring)

Loading