> I try to build a model for five parameters
> But, We found these five parameters have multicollinearity. We observed a
significant correlation between these parameters. So, we performed a PCA to
convert the set of five correlated air pollution variables into a set of
linearly uncorrelated main variations. (PC1 and PC2)> Then, we build the new GAM model using these two main variations as
predictor variables.> Model = gam( Y ~ s(PC1) + s(PC2))
This question is probably off-topic.
Which may explain why no one else has answered.
I think what you are trying to do, is contrary to the way that GAMs (and
regression generally) are designed to work.
I'd recommend that you use one approach or the other (PCA or GAMs, but not
both) unless you really know what you're doing...
And even if you do know what you're doing, the resulting model would be
difficult to interpret.
[[alternative HTML version deleted]]