Computes the Inverse Probability of Censoring Weighted (IPCW) estimator for the mean number of recurrent events. Supports various estimators including the Ghosh-Lin and Lawless-Cook estimators.
recregIPCW(
formula,
data = data,
cause = 1,
cens.code = 0,
death.code = 2,
cens.model = ~1,
km = TRUE,
times = NULL,
beta = NULL,
offset = NULL,
type = c("II", "I"),
marks = NULL,
weights = NULL,
model = c("exp", "lin"),
no.opt = FALSE,
augmentation = NULL,
method = "nr",
se = TRUE,
...
)An object of class "binreg" (extending "resmean") containing:
Estimated coefficients.
Variance-covariance matrix.
Influence functions.
Time points.
Observed IPCW weighted counts (increment IPCW) $$\int_0^t (1/G_c(s)) dN(s)$$.
Formula with an 'Event' outcome.
Data frame.
Cause of interest (default is 1).
Censoring code (default is 0).
Death code (default is 2).
Formula for the censoring model (default is ~1).
Logical; if TRUE, uses Kaplan-Meier for censoring weights; otherwise uses Cox model.
Time points for estimation (required).
Initial values for coefficients (optional).
Offsets.
Type of estimator: "II" (default) or "I".
Mark values.
Weights.
Model type for the mean: "exp" (default) or "lin".
Logical; if TRUE, skips optimization.
Augmentation terms.
Optimization method (default is "nr").
Logical; if TRUE, computes standard errors.
Additional arguments.
Thomas Scheike
recreg