R help - Oct 2012 - predict glm() with offset

Dear R friends.
I have a question about running a glm( family= 'binomial', *offset=T*),
(I
know offset is a vector of values)
My doubt is about predicting the values on a new data. Does the predict()
function considers the offset? o should I especified something?

Here is the model I´m using:

*model<-stepAIC(glm(f_ocur~altitud+UTM_X+UTM_Y+j_sin+j_cos+temp_res+pp,
offset=(log(1/offset)) data=data,
family='binomial', ), direction='both')*

Now, I want to get the estimated values, how should I do it considering
that I´m stablishing an offset value in the model?

If anyone can help me, i´d really appreciate it.

Thank you.

Lucas.

	[[alternative HTML version deleted]]