EventData: Constructor for event data object

In scientific-computing-solutions/eventPrediction: Event Prediction in Clinical Trials with Time-to-Event Outcomes

Description

All dates must be in one of the following formats: YYYY-MM-DD, DD/MM/YY or DD Month YYYY

Usage

EventData(data, subject, rand.date, has.event, withdrawn, time, site = NULL,
  event.type = NULL, remove.0.time = FALSE, followup = Inf)

Arguments

data	A data frame
subject	string, the column name of subject identifiers
rand.date	string, the column name of randomization dates
has.event	string, the column name of whether a subject had an event (1) or not (0)
withdrawn	string, the column name of whether a subject has withdrawn (1) or not (0)
time	Either a string, the column name of time each subject has been on the study or a list with elements with (some of) the following named elements. last.date, event.date, prog.date, dth.date and withdrawn.date In this case the package will attempt to derive the time column See the vignette and then eventPrediction:::AddTimeColumn for further details
site	optional column for subject site
event.type	optional column for the event type (e.g. unstable angina) if not included then ‘Had Event’ will be used
remove.0.time	logical, if TRUE then all subjects with time = NA or 0 are removed from the data set and not included in the object. If FALSE then they are included in the simulation (but not in the model fitting)
followup	A numeric value for the fixed follow up period a subject is followed. This is in days. If there is no fixed followup period then Inf should be used

Value

An EventData object

scientific-computing-solutions/eventPrediction documentation built on May 29, 2019, 3:44 p.m.