Displaying 1 result from an estimated 1 matches for "factorfutz".
2005 Nov 17
1
Predicting and Plotting "hypothetical" values of factors
...n, then it is tough to plot the S-shaped curve that describes
the probabilities.
Here is some code with comments describing the difficulties that we have
found. Are there simpler, less frustrating approaches?
-----------------------------
# Paul Johnson <pauljohn at ku.edu> 2005-11-17
# factorFutzing-1.R
myfac <- factor(c(1.1, 4.5, 1.1, 1.1, 4.5, 1.1, 4.5, 1.1))
y <- c(0,1,0,1,0,0,1,0)
mymod1 <-glm (y~myfac, family=binomial)
p1 <- predict(mymod1, type="response")
plot(myfac,p1)
# Wait, that's not the picture I wanted. I want to see the
# proverbial S-shaped c...