survival (version 2.11-4)

predict.coxph: Predictions for Cox model

Description

Compute fitted values and regression terms for a model fitted by coxph

Usage

predict.coxph(object, newdata, type=c("lp", "risk", "expected", "terms"), se.fit=FALSE, terms=names(object$assign), collapse, safe=FALSE,...)

Arguments

object

A coxph object

newdata

Optional new data to predict at

type

Type of prediction wanted

se.fit

Return standard errors as well?

terms

If type="terms", which terms to return.

collapse

identifier for which rows correspond to different individuals

safe

Use a more intensive `safe' prediction method

...

other unused arguments

Value

• A vector or matrix of fitted values. If se.fit=TRUE a list whose first component is the fitted values and second is the standard errors.

See Also

predict,coxph,termplot

Examples

data(aml)
fit<-coxph(Surv(time,status)~x,data=aml)
predict(fit,type="lp")
predict(fit,type="risk")
predict(fit,type="expected")
predict(fit,type="terms")
predict(fit,type="lp",se.fit=TRUE)
predict(fit,type="risk",se.fit=TRUE)
predict(fit,type="expected",se.fit=TRUE)
predict(fit,type="terms",se.fit=TRUE)