seqsurv: Generate a survfit object for state survival times.

Description

The function considers the spells in the different states in sequences and fits survival curves for each state. Alternatively, for a selected state, it fits the survival curves for each level of a stratifying group variable.

Survival curves are fitted with the survfit function.

Usage

seqsurv(seqdata, groups = NULL, per.state = FALSE, state = NULL,
    with.missing = FALSE)

Arguments

seqdata

A sequence stslist object as defined by the seqdef function.

groups

A stratifying group variable of length equal to the number of sequences.

per.state

Logical. Should the survival probabilites be computed for the state specified as state argument? If set as TRUE, the state argument must also be specified.

state

Single state value or a vector. The short name of the state for which to compute survival probabilities. If a vector of state names, survival probabilities are computed for the subset defined by those states. If NULL, survival probabilities are computed for all cases.

with.missing

Logical. Should the missing state be accounted for? (Not yet implemented!)

Value

An object of class stslist.surv. There is a plot method for such objects.

Details

The function considers the spells in the different states of a state sequence object (of class stslist).

When per.state = FALSE, it fits survival curves for each state in the alphabet. Currently, per.state = FALSE cannot be used with a non-NULL groups argument. However, seqsplot handles this case.

When per.state = TRUE, the survival curve is fitted only for the state provided as state argument. This is done for each level of the groups variable.

Survival curves are fitted with the survfit function.

Examples

Run this code

# NOT RUN {
## Defining a sequence object with the data in columns 10 to 25
## (family status from age 15 to 30) in the biofam data set
data(biofam)
biofam.lab <- c("Parent", "Left", "Married", "Left+Marr",
                "Child", "Left+Child", "Left+Marr+Child", "Divorced")
biofam.short <- c("P","L","M","LM","C","LC","LMC","D")

sple <- 500:700  ## want a sample with all elements of the alphabet
##seqstatl(biofam[sple,10:25])
biofam <- biofam[sple,]

## creating the state sequence object
biofam.seq <- seqdef(biofam[,10:25], alphabet=0:7, states=biofam.short, labels=biofam.lab)

## Spell survival curves
(biofam.surv <- seqsurv(biofam.seq))

## Cohort distinguishing between those born before or after World War II
biofam$wwii <- biofam$birthyr <= 1945

## Separate survival curves in a given state (here LMC "Left+Marr+Child") according to wwii
(biofam.surv <- seqsurv(biofam.seq, groups=biofam$wwii, per.state=TRUE, state="LMC"))
plot(biofam.surv)
# }