The trace of the hat matrix corresponds to the effective degrees of freedom (edf) of a generalized additive model. The edf is an internal measure of model complexity.
calcTrAFast(X, w, lambda=0)Design matrix of the covariates.
Diagonal weight matrix of the pseudo iterated least squares algorithm. See function calcWdiag.
Regularization parameter of kernel ridge regression. Default is 0 (numeric scalar).
Effective degrees of freedom of a generalized additive model with regularization (numeric scalar).
This function is a more computational efficient version of calcTrA. The general algorithm is simplified, requires less memory and is faster. Therefore it is better suited for data sets above 1000 observations.
Simon N. Wood, (2006), Generalized Additive Models: An Introduction with R, Taylor \& Francis Group LLC