predict.cosso: Make predictions or extract coefficients from a cosso model

Description

Make prediction for future observations or extract the model parameters at a particular smoothing parameter.

Usage

# S3 method for cosso
predict(object,xnew,M,type=c("fit","coefficients","nonzero"),eps=1e-7,...)

Value

The object returned depends on type.

When type="fit", predicted $$eta$$ function value will be given at the new design points xnew.

When type="coefficients", three sets of coefficients will be returned.

Intercept: the estimated intercept. If family="Cox", the intercept is zero.
coefs: the estimated coefficients for kernel representers.
theta: the estimated scale parameters for each functional component.

When type="nonzero", a list of the indices of the nonconstant functional components will be returned.

Arguments

object: a cosso object.
xnew: matrix of new values for $x$ at which predictions are to be made. Object must be a matrix and have the same dimension as the training design points. Continuous variable will also have to be scaled to [0,1] interval.
M: a smoothing parameter value. M should be taken between 0 and p. If not provided, a cross-validation procedure will be carried out to select an appropriate value.
type: if type="fit" (default), fitted values will be returned. If type="coefficients", model coefficients will be returned. Abbreviations are allowed.
eps: an effective zero, default is 1e-7
...: additional arguments for predict function.

Author

Hao Helen Zhang and Chen-Yen Lin

Examples

Run this code

## Gaussian
set.seed(20130310)
x=cbind(rbinom(200,1,.7),matrix(runif(200*7,0,1),nc=7))
y=x[,1]+sin(2*pi*x[,2])+5*(x[,4]-0.4)^2+rnorm(200,0,1)
G.Obj=cosso(x,y,family="Gaussian")
predict.cosso(G.Obj,M=2,type="nonzero")
predict.cosso(G.Obj,xnew=x[1:3,],M=2,type="fit")
## Clean up
rm(list=ls())

if (FALSE) {
## Binomial
set.seed(20130310)
x=cbind(rbinom(200,1,.7),matrix(runif(200*9,0,1),nc=9))
trueProb=1/(1+exp(-x[,1]-sin(2*pi*x[,2])-5*(x[,4]-0.4)^2))
y=rbinom(200,1,trueProb)

B.Obj=cosso(x,y,family="Bin") 
f.hat=predict.cosso(B.Obj,xnew=x,M=2,type="fit")
prob.hat=1/(1+exp(-f.hat))
## Clean up
rm(list=ls())
}