lifeTable: Life Table

In discSurv: Discrete Time Survival Analysis

Life Table

Description

Constructs a life table and estimates discrete hazards, survival functions, discrete cumulative hazards and their standard errors without covariates.

Usage

lifeTable(
  dataShort,
  timeColumn,
  eventColumn,
  intervalLimits = NULL,
  censInterval = "middle"
)

## S3 method for class 'discSurvLifeTable'
print(x, firstRows = 5, ...)

Arguments

dataShort	Original data in short format (class "data.frame").
timeColumn	Name of the column with discrete survival times (class "character").
eventColumn	Gives the column name of the event indicator (1=observed, 0=censored) (class "character").
intervalLimits	Optional names of the intervals for each row, e. g. \left[ a_0, a_1 \right), \left[ a_1, a_2 \right), ..., \left[ a_{q-1}, a_q \right) (class "character")
censInterval	Assumption about when censoring takes places within an interval on the continuous time scale (class "character"). Possible values are "start", "middle", "end". Default is "middle".
x	estimated life table (class "discSurvLifeTable")
firstRows	Display the first number of specified rows (class "numeric").
...	Specification of additional arguments in function print.

Details

The assumption of censoring times within the given intervals of argument "censInterval" estimate the hazard rate with value "start" $d_t/(n_t-w_t)$, "middle" $d_t/(n_t-w_t/2)$ and "end" $d_t/n_t$

Value

List containing an object of class "data.frame" with following columns

• n Number of individuals at risk in a given time interval (integer)

• events Observed number of events in a given time interval (integer)

• dropouts Observed number of dropouts in a given time interval (integer)

• atRisk Estimated number of individuals at risk, corrected by dropouts (numeric)

• hazard Estimated risk of death (without covariates) in a given time interval

• seHazard Estimated standard deviation of estimated hazard

• S Estimated survival curve

• seS Estimated standard deviation of estimated survival function

• cumHazard Estimated cumulative hazard function

• seCumHazard Estimated standard deviation of the estimated cumulative hazard function

• margProb Estimated marginal probability of event in time interval

Author(s)

Thomas Welchowski [e-mail]

Matthias Schmid [e-mail]

References

\insertRef

lawlessLifetimediscSurv

\insertReftutzModelDiscdiscSurv

Examples

# Example with unemployment data
library(Ecdat)
data(UnempDur)

# Extract subset of all persons smaller or equal the median of age
UnempDurSubset <- subset(UnempDur, age <= median(UnempDur$age))
LifeTabUnempDur <- lifeTable(dataShort = UnempDurSubset, timeColumn = "spell", 
eventColumn = "censor1")
LifeTabUnempDur

# Example with monoclonal gammapothy data
library(survival)
head(mgus)

# Extract subset of mgus
subMgus <- mgus [mgus$futime<=median(mgus$futime), ]

# Transform time in days to intervals [0, 1), [1, 2), [2, 3), ... , [12460, 12461)
mgusInt <- subMgus
mgusInt$futime <- mgusInt$futime + 1
LifeTabGamma <- lifeTable(dataShort = mgusInt, timeColumn= "futime", eventColumn = "death")
head(LifeTabGamma$Output, 25)
plot(x = 1:dim(LifeTabGamma$Output)[1], y = LifeTabGamma$Output$hazard, type = "l", 
xlab = "Time interval", ylab = "Hazard", las = 1, 
main = "Life table estimated marginal discrete hazards")

discSurv documentation built on April 29, 2026, 9:07 a.m.