[R] Discrete variables with logistf

smb123 Shannon.Bros at sjsu.edu
Tue Feb 17 17:59:46 CET 2015


I am relatively new to R and I would appreciate some advice on how to include
both discrete and continuous predictor variables in a stepwise backward log
likelihood ration logistic regression. I have a model that includes 3
continuous predictor variables (soilmois, grasscov, and ranvar) and one
discrete predictor (grnsqrl) coded as 0 and 1.

The script I used was:
Bdata$grnsqrl <- factor(Bdata$grnsqrl)
fit2 <-logistf(data=Bdata, badger~soilmois+grasscov+ranvar+grnsqrl)
summary(fit2)
backward(fit2)
drop1(fit2)

The model summary works fine but for either backward.logistf or
drop1.logistf give the error:
missing value where TRUE/FALSE needed

Thank you!



--
View this message in context: http://r.789695.n4.nabble.com/Discrete-variables-with-logistf-tp4703403.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list