This part of the chapter describes some features which may be interesting for special problems.
The gam quantlib automatically loads the quantlibs xplore , glm and gplm , if not yet active.
The quantlets
gintestpl
and
intestpl
perform the estimation on an
internal grid if the number of observations exceeds 50 and 40,
respectively. They interpolate the estimated additive functions for the
explanatory variable t, or, if given, for the grid tg.
For graphical output
gamfit
makes use of the auxiliary quantlet
gamout
.
|
library("gam") n = 100 v =uniform(n,4) x =v[,2:4] y =x[,1]^2+0.1*x[,2]+normal(n) h =0.5 bn=0.02 gest=pcad(x,x,y,h,bn) gest.jhat gest.g gest.mhat
The quantlet
pcad
estimates the additive components, the significant
directions and the regression function on principal components. It
accepts only one-dimensional y. The standard call is:
Input parameters:
The quantlet returns