pact.cv: Cross-validation for pact

Description Usage Arguments Details Value Author(s) Examples

View source: R/pact.R

Description

Predictive scores using k-fold cross-validation for the model developed in pact.fit

Usage

1
pact.cv(p, nfold)

Arguments

p

An object of class 'pact'

nfold

The number of folds (k) for the k-fold cross-validation. k equal to the sample size would mean a leave-one-out cross-validation

Details

Obtain cross-validated predictive scores for the model developed in pact.fit. In each fold of the cross-validation, a model is developed from the observations in the training set using the same variable selection parameters as that used for the model developed in pact.fit. The estimated coefficients of the regression model developed using training set are used to make predictions for the left out observations (test set). This is repeated for all the folds. Scores are thus obtained for all the subjects in the dataset. The function eval.pact.cv provides various evaluation options for the cross-validated scores.

Value

A list with the following components

PredScore

The cross-validated scores for each subject (a vector)

Y

The response variable used

Xf

The dataframe of fixed prognostic covariates

Xv

The dataframe of candidate predictive variables

Treatment

The treatment assignment indicator used

nCovarf

The number of variables in Xf

nCovarv

The number of variables in Xv

family

Type of the response variable

varSelect

The variable selection method used

nsig, cvfolds.varSelect, which.lambda, penalty.scaling

The variable selection parameters used

call

The call that produced this output

Author(s)

Jyothi Subramanian and Richard Simon
Maintainer: Jyothi Subramanian <[email protected]>

Examples

1
2
3
4
5
6
7
data(prostateCancer)
Y <- prostateCancer[,3:4]
Xf <- prostateCancer[,7:8]
Xv <- prostateCancer[,c(5:6,9)]
Treatment <- as.factor(prostateCancer[,2])
p <- pact.fit(Y=Y,Xf=Xf,Xv=Xv,Treatment=Treatment,family="cox",varSelect="lasso")
cv <- pact.cv(p, nfold=5)

pact documentation built on May 29, 2017, 3:12 p.m.