pew {dynpred}

Calculate dynamic prediction error curve

Description

Calculate dynamic fixed width prediction error curve.

Usage

pew(time, status, tsurv, survmat, tcens, censmat, width, FUN = c("KL",
  "Brier"), tout)

pewcox(formula, censformula, width, data, censdata, FUN = c("KL", "Brier"),
  tout, CV = FALSE, progress = FALSE)

Arguments

time

Vector of time points in data

status

Vector of event indicators in data

tsurv

Vector of time points corresponding to the estimated survival probabilities in survmat

survmat

Matrix of estimated survival probabilities; dimension should be length of tsurv x length of time

tcens

Vector of time points corresponding to the estimated censoring probabilities in censmat

censmat

Matrix of estimated censoring probabilities; dimension should be length of tcens x length of time

width

Width of the window

FUN

The error function, either "KL" (default) for Kullback-Leibler or "Brier" for Brier score

tout

Vector of time points at which to evaluate prediction error. If missing, prediction error will be evaluated at all time points where the estimate will change value

formula

Formula for prediction model to be used as in coxph

censformula

Formula for censoring model, also to be used as in coxph

data

Data set in which to interpret formula

censdata

Data set in which to interpret censformula

CV

Boolean (default=FALSE); if TRUE, (leave-one-out) cross-validation is used for the survival probabilities

progress

Boolean (default=FALSE); if TRUE, progress is printed on screen

Details

Corresponds to Equation (3.6) in van Houwelingen and Putter (2011). The censformula is used to calculate inverse probability of censoring weights (IPCW).

Value

A data frame with columns

time

Event time points

Err

Prediction error of model specified by formula at these time points

and with attribute "width" given as input.

Author(s)

Hein Putter [e-mail]

References

van Houwelingen HC, Putter H (2012). Dynamic Prediction in Clinical Survival Analysis. Chapman & Hall.

Examples

data(ova)
# Example on a subset, because the effect of CV is clearer
ova2 <- ova[1:100,]
pewcox(Surv(tyears, d) ~ Karn + Broders + FIGO + Ascites + Diam, Surv(tyears, 1-d) ~ 1,
  width=2, data = ova2, FUN="Brier", tout=seq(0,6,by=0.5))
pewcox(Surv(tyears, d) ~ Karn + Broders + FIGO + Ascites + Diam, Surv(tyears, 1-d) ~ 1,
  width=2, data = ova2, FUN="Brier", tout=seq(0,6,by=0.5), CV=TRUE, progress=TRUE)


pewcox(Surv(tyears, d) ~ Karn + Broders + FIGO + Ascites + Diam, Surv(tyears, 1-d) ~ 1,
  width=2, data = ova, FUN="Brier", tout=seq(0,6,by=0.5))
pewcox(Surv(tyears, d) ~ Karn + Broders + FIGO + Ascites + Diam, Surv(tyears, 1-d) ~ 1,
  width=2, data = ova, FUN="Brier", tout=seq(0,6,by=0.5), CV=TRUE, progress=TRUE)