ps: Specify a smoothing spline fit in a PSgam formula

In DPpackage: Bayesian Nonparametric Modeling in R

Description

A symbolic wrapper to indicate a smooth term in a formula argument to PSgam

Usage

ps(..., k=50,degree=3,pord=1)

Arguments

...	the predictors.
k	an integer giving the number of knots. The number of basis functions is k+degree.
degree	an integer giving the degree of the B-splines, e.g. degree=3 gives a cubic spline.
pord	an integer giving the order of difference penalty.

Value

ps returns the vectors of predictors, endowed with a number of attributes. The vector itself is used in the construction of the model matrix.

Note that ps does not do the smoothing; it simply sets things up for PSgam.

Author(s)

Alejandro Jara <[e-mail]>

References

Eilers, P.H.C. and Marx, B.D. (1996) Flexible Smoothing with B-splines and penalties. Statistical Science, 11(2): 89-121.

Examples

   # fit Start using a smoothing spline with 4 df.
     y ~ Age + ps(Start, degree=4)
   # fit log(Start) using a smoothing spline with 5 df.
     y ~ Age + ps(log(Start), degree=5)

DPpackage documentation built on May 1, 2019, 10:23 p.m.