Learn R Programming
⚠️There's a newer version (2023.04.12) of this package.Take me there.

pec (version 2018.07.26)

Prediction Error Curves for Risk Prediction Models in Survival Analysis

Description

Validation of risk predictions obtained from survival models and competing risk models based on censored data using inverse weighting and cross-validation.

Copy Link

Version

Install

install.packages('pec')

Monthly Downloads

7,095

Version

2018.07.26

License

GPL (>= 2)

Maintainer

Thomas Alexander Gerds

Last Published

July 26th, 2018

Functions in pec (2018.07.26)

R2

Explained variation for survival models
GBSG2

German Breast Cancer Study Group 2
cost

Copenhagen Stroke Study
coxboost

Formula interface for function CoxBoost of package CoxBoost.
Pbc3

Pbc3 data
plot.calibrationPlot

Plot objects obtained with calPlot
plot.pec

Plotting prediction error curves
pecCforest

S3-wrapper function for cforest from the party package
plotPredictEventProb

Plotting predicted survival curves.
pec

Prediction error curves
calPlot

Calibration plots for right censored data
plotPredictSurvProb

Plotting predicted survival curves.
cindex

Concordance index for right censored survival time data
resolvesplitMethod

Resolve the splitMethod for estimation of prediction performance
selectCox

Backward variable selection in the Cox regression model
predictRestrictedMeanTime

Predicting restricted mean time
predictSurvProb

Predicting survival probabilities
ipcw

Estimation of censoring probabilities
crps

Summarizing prediction error curves
predictLifeYearsLost

Predicting life years lost (cumulative cumulative incidences) in competing risk models.
predictEventProb

Predicting event probabilities (cumulative incidences) in competing risk models.
threecity

threecity data
reclass

Retrospective risk reclassification table
print.pec

Printing a `pec' (prediction error curve) object.
pecCtree

S3-Wrapper for ctree.
pecRpart

Predict survival based on rpart tree object
selectFGR

Stepwise variable selection in the Fine & Gray regression competing risk model
simCost

Simulate COST alike data
Special

Drawing bootstrapped cross-validation curves and the .632 or .632plus error of models. The prediction error for an optional benchmark model can be added together with bootstrapped cross-validation error and apparent errors.