| Title: | Compare C-Statistics with Bootstrapped Confidence Intervals |
|---|---|
| Description: | Provides a function to compare C-statistics (concordance statistics) between two survival models using bootstrap resampling. Returns confidence intervals and a p-value for the difference in C-statistics. Useful for model evaluation and comparison. |
| Authors: | Hairong Liu [aut], Ning Meng [aut, cre] |
| Maintainer: | Ning Meng <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.1.0 |
| Built: | 2026-06-02 07:42:38 UTC |
| Source: | https://github.com/lemonade0924/comparecstat |
This function compares the C-statistics of two fitted models using bootstrap resampling.
compare_c_stat(model_raw, model_ext, data, R = 1000, ci_type = "perc")compare_c_stat(model_raw, model_ext, data, R = 1000, ci_type = "perc")
model_raw |
A fitted model (e.g., coxph) representing the base model. |
model_ext |
A fitted model (e.g., coxph) representing the extended model. |
data |
The dataset used for fitting the models. |
R |
Number of bootstrap replications. Default is 1000. |
ci_type |
Type of confidence interval to return ("perc", "norm", "basic", etc.). |
A data frame showing C-statistics for each model, their bootstrapped confidence intervals, and the p-value for the difference.
library(survival) data(lung) model1 <- coxph(Surv(time, status) ~ age, data = lung) model2 <- coxph(Surv(time, status) ~ age + sex, data = lung) compare_c_stat(model1, model2, data = lung, R = 500)library(survival) data(lung) model1 <- coxph(Surv(time, status) ~ age, data = lung) model2 <- coxph(Surv(time, status) ~ age + sex, data = lung) compare_c_stat(model1, model2, data = lung, R = 500)