[R] endogenous variables in gam (mgcv)

[Michael A. Milligan]

Hello,

I have a question.  Suppose that I have a function to estimate with gam (in the mgcv package),
y=s(x1)+s(x2)+XB
where X is a vector of exogenous variables and x1 and x2 are explanatory variables assumed parametric linear functions of X and other exogenous variables Z.  Is there a way to evaluate this equation with gam, allowing for endogeneity?  If not, is there another semiparametric/nonparametric evaluation function where endogeneity can be programed more easily?

I really appreciate any help I can get with this.  Thank you very much.

Michael Milligan
Doctoral Candidate
University of New Mexico