generate_bounds: Generates the intervals based on the survival times in the...

In spect: Survival Prediction Ensemble Classification Tool

Generates the intervals based on the survival times in the supplied data set using the quantile function.

Description

Generates the intervals based on the survival times in the supplied data set using the quantile function.

Usage

generate_bounds(
  train_data,
  time_col,
  event_col,
  suggested_intervals,
  obs_window
)

Arguments

train_data	A survival data set containing at least three columns - one which matches the string in the 'time_col' parameter, one which matches the string in the 'event_col' parameter, and at least one covariate column for modeling.
time_col	The name of the column in 'train_data' containing survivial time
event_col	The name of the column in 'train_data' contaiing the event indicator. Values in this column must be either zero (0) or one (1)
suggested_intervals	The number of intervals to create. If the number of events in the data is less than 'suggested_intervals', it is ignored.
obs_window	An artificial censoring time. Any observations in 'train_data' beyond this time will be administratively censored.

Value

A list of upper an lower bounds for each generated interval.

Author(s)

Stephen Abrams, [e-mail]

See Also

[create_person_period_data()]

Examples

df <- data.frame(a=c(1,2,3,4,5,6), surv_time=c(1,4,5,6,8,9), event=c(1,1,1,1,0,1))
bounds <- generate_bounds(df, time_col="surv_time", event_col="event", 
                          suggested_intervals=3, obs_window=8)
                          

spect documentation built on April 11, 2025, 6:17 p.m.