Predict probabilities from a survfit object — predict.survfit • causalRisk

Function to predict probabilities from a survfit object

# S3 method for survfit
predict(
  object,
  newdata = NULL,
  timevar = rlang::sym(".Ymin"),
  type = "surv",
  times = NULL,
  ...
)

Arguments

object: the results of a survfit fit
newdata: Optional new data at which to do predictions. If absent predictions are for the data frame used in the original fit
timevar: Name of time variable in new data
type: The type of predicted value. Options are linear predictor "lp" (which is always 0), the risk score exp(lp) "risk" (which is always 1), the cumulative hazard "cumhaz", the survival probability "surv", or the expected number of events "expected" (-log(surv))
times: The times at which to perform predictions
...: Unused

Alexander Breskin