scam provides functions for generalized additive modelling under shape constraints on the component functions of the linear predictor of the GAM. Models can contain multiple shape constrained and unconstrained terms as well as bivariate smooths with double or single monotonicity. The model set up is the same as in mgcv(gam) with the added
shape constrained smooths, so the unconstrained smooths can be of more than one variable, and other
user defined smooths can be included. Penalized log likelihood maximization is used to fit the model
together with the automatic smoothness selection.scam and plot.scam. These functions are based on the functions of the
unconstrained GAM mgcv(gam) and mgcv(plot.gam) and similar in use.
summary.scam allows to extract the results of the model fitting in the same way
as in summary.gam. A Bayesian approach is used to obtain a covariance matrix of the model
coefficients and credible intervals for each smooth.