MAPLES (version 1.0)

epdata: Prepares episode data for event history analysis.

Description

Starting from a micro-level dataset containing dates for some relevant events, it prepares a file with an episode-data format suitable for event history analysis.

Usage

epdata(start, event, lcensor, rcensor,  subset = TRUE, birth, id, addvar)

Arguments

start

a vector of dates at which the unit starts to be at risk to experience the transition. It may be unknown in case of left-censoring. When the episode is no left censored, start coincides with the beginning of the episode.

event

a vector of containing the date of the event of interest. It may be unknown in case of right-censoring.

lcensor

an optional vector of left censoring dates.

rcensor

an optional vector of right censoring dates.

subset

a logical expression indicating elements or rows to keep. It specifies the subset of cases at risk of experiencing the event of interest. If TRUE, the whole sample will be considered.

birth

a vector containing the respondent date of birth. This information in useful in order to compute ages at various events (NA not allowed).

id

a vector of identification numbers. If id remains unspecified, a default id is implemented.

addvar

a dataframe containing additional variables to add in the resulting output.

Value

'counting process' formulation of survival analysis, each record ha the following variables:

Tstart

episode starting date

Tstop

episode ending date

status

0: right censored: 1: event occurred: 2: left censored: 3: interval censored. Is equal to 1 if and only if the date of the event precedes the date of right censoring.

Agestart

Age at starting date (Tstart)

Agestop

Age at the ending date (Tstop)

Details

All the specified objects must have the same length (i.e. the number of cases must be the same). For any observation at least one between start and lcensor and one between event and rcensor must be known. In general, we have start

References

Impicciatore R. and Billari F.C., (2010), MAPLES: a general method for the estimation of age profiles from standard demographic surveys (with an application to fertility), DEAS WP, http://ideas.repec.org/p/mil/wpdepa/2010-40.html

See Also

splitter, ageprofile, plotap

Examples

# creates an episode-data structure relating 
# to the transition childless-->first child
ep1<-epdata(start=demogr$dbirth, 
            event=demogr$dch1,
            rcensor=demogr$dint,
            birth=demogr$dbirth,
            id=demogr$id,
            addvar=subset(demogr,select=c(-id,-dbirth)))    
                 
# or, using 'with'
ep1<-with(demogr,epdata(start=dbirth, event=dch1, 
          rcensor=dint, birth=dbirth,id=id,
          addvar=subset(demogr,select=c(-id,-dbirth))))