[R] comparison between parametric and semiparametric models

[Wensui Liu]

Dear Listers,

I have a question somehow off-topic but think I can find a good answer here.

Using the same dataset, I've built a logistic regression and
generalized additive model. Might I use likelihood ratio test to see
if there is significant improvement in GAM model? Something like:
(deviance of GLM - deviance of GAM) with degree of free = additional
number of DF in GAM using chisquare test.

Thank you so much!