predictRisk.pcoxtime: Extract predictions from pcoxtime model

In CYGUBICKO/pcoxtime-pkg: Penalized Cox Proportional Hazard Model for Time-Dependent Covariates

Extract predictions from pcoxtime model

Description

Extract event probabilities from the fitted model.

Usage

## S3 method for class 'pcoxtime'
predictRisk(object, newdata, times, ...)

Arguments

object	fitted pcoxtime.
newdata	a data frame containing the variables appearing on the right hand side of pcoxtime formula.
times	a vector of times in the range of the response, at which to return the survival probabilities.
...	for future implementations.

Details

For survival outcome, the function predicts the risk, 1 - S(t|x), where S(t|x) is the survival chance of an individual characterized by x.

Value

a matrix of probabilities with as many rows as the rows of the newdata and as many columns as number of time points (times).

Examples

if (packageVersion("survival")>="3.2.9") {
   data(cancer, package="survival")
} else {
   data(veteran, package="survival")
}
# Penalized
lam <- 0.1
alp <- 0.5
pfit1 <- pcoxtime(Surv(time, status) ~ factor(trt) + karno + diagtime + age + prior
	, data = veteran
	, lambda = lam
	, alpha = alp
)
r1 <- predictRisk(pfit1, newdata = veteran[1:80,], times = 10)

# Unpenalized
lam <- 0
alp <- 1
pfit2 <- pcoxtime(Surv(time, status) ~ factor(trt) + karno + diagtime + age + prior
	, data = veteran
	, lambda = lam
	, alpha = alp
)
r2 <- predictRisk(pfit2, newdata = veteran[1:80,], times = 10)
plot(r1, r2, xlim=c(0,1), ylim=c(0,1)
	, xlab = "Penalized predicted survival chance at 10"
	, ylab="Unpenalized predicted survival chance at 10"
)

CYGUBICKO/pcoxtime-pkg documentation built on Aug. 5, 2023, 1:35 p.m.