Dear All, I was asked to criticize the following strategy for logistic
regression model building
and I would appreciate any comments from the list (positive or negative):
1. After exclusion of variables based on subject knowledge, 25 variables were
considered
as possible candidates (sample size ~200, with smaller group~ 90)
2. An all-subsets regression technique was employed and the model chosen was
the one
with the largest bias-corrected (bootstrap, B=200) ROC curve area
3. ?To check stability of the model?, another bootstrap procedure was performed
(B=200
again) and step 2 was included in every bootstrap sample (? Double bootstrap).
So, each
bootstrap resample had its ?best model? with its associated optimism (obtained
from step
2).
4. The number of times that each variable appeared in the 200 bootstrap samples
(200
?best models?, but not necessarily all different) from step 3 was reported.
5. The final model reported was the one obtained after all-subsets regression
was applied
to the original sample, and although the variables in this model were the ones
who showed
the highest frequency in the 200 bootstrap samples, it was recognized that
several other
"best models" were possible, which was illustrated by the frequency of
variables in the 200
bootstrap resamples. The ?optimism? was reported as the average of the 200
?optimisms?.
Thanks in advance,
Fernando Medeiros.
Fernando Medeiros, M.D., Ph.D.
Department of Neurosciences
University of Sao Paolo
---
Acabe com aquelas janelinhas que pulam na sua tela.
AntiPop-up UOL - É grátis!
http://antipopup.uol.com.br
|