Displaying 1 result from an estimated 1 matches for "smokentype1".
Did you mean:
smokentype2
2011 Jun 13
0
How to formulate an (effect-modifying) interaction with matching variable in a conditional logistic regression?
...ore unwieldy with factors as matching variables
# I fake a second matching variable "type" with three levels
dataset$type<-c(rep(1,36),rep(2,38),rep(3,38))
dataset$type<-as.factor(dataset$type)
levels(dataset$type)<-c("type1","type2","type3")
dataset$smokeNtype1<-as.numeric(dataset$smoke=="yes")*as.numeric(dataset$type=="type1")
dataset$smokeNtype2<-as.numeric(dataset$smoke=="yes")*as.numeric(dataset$type=="type2")
multi4.cl<-update(mult.cl,.~.+smokeNtype1+smokeNtype2)