summarizeHazard: Convert hazard to survival function and instantaneous event...

In WinVector/QSurvival: Quick survival models by data transformation

Description

Sums up hazard function grouped by original ID variable ordered by step.

Usage

summarizeHazard(d, idColumnName, indexColumnName, hazardColumnNames,
  survivalColumnNames = "survival", deathIntensityColumnNames = NULL,
  parallelCluster = NULL)

Arguments

d	data.frame
idColumnName	character scalar, column containing original row ids
indexColumnName	character scalar, column containing quasi event indices. Indices must be complete intevvals 1:k for some k.
hazardColumnNames	character vector, columns containing hazard scores
survivalColumnNames	character vector, columns to write survival probability in
deathIntensityColumnNames	vector scalar, columns to write death intensities
parallelCluster	optional parallel cluster to run on

Value

list, details=survival data.frame, expectedLifetime has lifetime estimates (lifetime seen in the scoring windows plus a geometric term to count time steps past the window).

Examples

d <- data.frame(lifetime=c(2,1,2),censored=c(FALSE,FALSE,TRUE))
d2 <- buildQuasiObsForComparison(d,5,d$lifetime,ifelse(d$censored,NA,d$lifetime),
   'origRow','sampleAge','deathEvent')
d2$hazardPred <- 0.1
summarizeHazard(d2,'origRow','sampleAge','hazardPred',
   survivalColumnName='survival',
   deathIntensityColumnName='deathIntensity')

WinVector/QSurvival documentation built on May 9, 2019, 10:59 p.m.