Fit an IPCW-weighted Cox proportional hazards model

Fits a weighted Cox model with a robust sandwich variance and returns log-hazard ratio estimates together with both model-based and robust standard errors, plus the hazard ratio and 95% confidence interval based on the robust standard error. Convenience wrapper for survival::coxph().

Usage

get_ipcw_cox_fit_se(data, covariate = "x", weight = "wgt")

Arguments

data

A data frame in long (counting-process) format, as returned by get_ipcw_wgt_se(). Must contain columns tstart, tstop, delta, id, the covariate named by covariate, and the weight column named by weight.

covariate

Character string. Name of the predictor covariate column. Default is "x".

weight

Character string. Name of the weight column. Default is "wgt".

Value

A data frame with one row per term containing:

term

Covariate name.

log_hr

Log hazard ratio estimate.

log_hr_se

Model-based standard error of the log hazard ratio.

log_hr_rob_se

Robust (sandwich) standard error of the log hazard ratio.

hr

Hazard ratio estimate.

hr_ci_low

Lower 95% confidence limit (based on robust SE).

hr_ci_high

Upper 95% confidence limit (based on robust SE).

Examples

set.seed(20240429)
dat <- sim_data_se(n = 500)
dat_long <- get_ipcw_wgt_se(dat)
get_ipcw_cox_fit_se(dat_long, weight = "wgt")
#> # A tibble: 1 × 7
#>   term  log_hr log_hr_se log_hr_rob_se    hr hr_ci_low hr_ci_high
#>   <chr>  <dbl>     <dbl>         <dbl> <dbl>     <dbl>      <dbl>
#> 1 x      -1.21     0.103         0.162 0.298     0.217      0.410