estimate_censoring_score_cox: Estimate Censoring Scores Using Cox Regression

Description

Estimate Censoring Scores Using Cox Regression

Usage

estimate_censoring_score_cox(
  data,
  time_var,
  treatment_var,
  formula,
  control = list(),
  ties = "efron"
)

Value

List with class "censoring_score_cox":

censoring_models: Fitted coxph objects by treatment level.
censoring_scores: P(C >= T_i | Z_i, X_i) for observed treatment.
censoring_matrix: (n x J) matrix of P(C >= T_i | Z=j, X_i).
n_levels: Number of treatment levels.
treatment_levels: Sorted treatment values.
model_type: "cox".
baseline_hazards: Baseline cumulative hazards by treatment level.
coef_list: Coefficient vectors by treatment level.
vcov_list: Variance-covariance matrices by treatment level.
linear_predictors_matrix: (n x J) matrix of linear predictors.

Arguments

data: Data frame.
time_var: Name of time variable.
treatment_var: Name of treatment variable.
formula: Censoring model formula. Use Surv(time, censor_indicator) ~ X1 + X2 where censor_indicator = 1 indicates censoring. If event is coded canonically (event=1, censored=0), use I(1-event). Otherwise, use the appropriate transformation. Treatment is automatically removed if included.
control: Control parameters for coxph(). Default list().
ties: Tie handling method. Default "efron".

Details

Fits Cox models within each treatment group. Censoring scores computed as: $$K_c^{(j)}(t, X) = \exp(-H_0^{(j)}(t) \cdot \exp(\beta_j' X))$$ where $H_0^{(j)}(t)$ is cumulative baseline hazard. Baseline hazards evaluated at nearest time point for each individual.