Applies a given dropout function to a simulated trial dataset, in order to create a "missing" variable representing subject dropout
Usage
createDropout(
data,
dropFunc,
seed = .deriveFromMasterSeed(),
idCol = getEctdColName("Subject"),
timeCol = getEctdColName("Time"),
flagName = getEctdColName("Missing"),
...
)Arguments
- data
(Required) Data frame to which to add a "missingness" flag
- dropFunc
(Required) Drop out function to apply to the data. The function must return a vector of booleans
- seed
(Optional) Random generation seed. By default, the current random seed is used
- idCol
(Optional) The name of the subject varuable in the data. "SUBJ" by default
- timeCol
(Optional) The name of the time column in the data. "TIME" by default
- flagName
(Optional) The name of the missing variable to create in the data. "MISSING" by default
- ...
(Optional) Extra arguments to be passed directly to the dropout function. No arguments are passed by default
Details
This function will accept (and check) a "dropout" function, which is to applied to the data. This function must return a boolean vector. The dropout function supplied will be applied to the data provided (possibly with extra arguments supplied with the ellipses). Once a boolean vector is returned, the function will ensure dropout is "retained" over time within subject (so if a subject drops out at visit 1, he/she is also missing at visits 2, 3 etc.). This "retained" boolean vector is added to the dataset (or used to enhance an existing "missing" flag) and the updated data is returned.
Author
Mike K Smith mstoolkit@googlemail.com
Examples
if (FALSE) {
dFun <- function(data, prop) sample(0:1, nrow(data), replace=TRUE, prob=c(1-prop, prop))
createDropout(data=myDf, dropFunc=dFun, prop=.05)
# SUBJ TIME MISSING
# 1 1 1 0
# 2 1 2 0
# 3 1 3 0
# 4 1 4 0
# 5 1 5 0
# 6 2 1 0
# 7 2 2 0
# 8 2 3 1
# 9 2 4 1
# 10 2 5 1
# 11 3 1 0
# 12 3 2 1
# 13 3 3 1
# 14 3 4 1
# 15 3 5 1
}