Skip to contents

Format of input dataset

Source: vignettes/Data_format.Rmd
Data_format.Rmd

SLCARE requires the input datset have a specific format. We will illustrate the format with the SLCARE_simdat from the SLCARE package.

library(SLCARE)
packageVersion("SLCARE")
#> [1] '1.0.0'
data("SLCARE_simdat", package = "SLCARE")
head(SLCARE_simdat)
#>       ID time event x1        x2
#> 1 UOM054   65     1  0 0.5555556
#> 2 UOM054  940     1  0 0.5555556
#> 3 UOM054 2420     1  0 0.5555556
#> 4 UOM054 2475     1  0 0.5555556
#> 5 UOM054 2650     1  0 0.5555556
#> 6 UOM054 3830     1  0 0.5555556

Dataset SLCARE_simdat consists of 5 columns. We can devide the 5 columns into 4 parts:

  • Column 1 ID: Subjects identification
  • Column 2 time: Time for recurrent event and censoring time (which usually represent the longest follow up time)
  • Column 3 event: Recurrent event indicator. A value of 1 represents the recurrent event is recorded; a value of 0 represents censoring.
  • Columns 4-5 x1 and x2: time independent corvariates.

Input datset should have the same format as SLCARE_simdat. Specifically, column 1-3 must be ID, time and event, and they must in the same order. Input datset can have as many corvariates as you need. They should be placed from column 4 which locates after ID, time and event.