How does this skill handle the Proportional Hazards assumption?

It includes specific patterns for testing the PH assumption using Schoenfeld residuals (cox.zph) and provides solutions for violations, such as stratified models or time-varying covariates.

Which R packages are supported by this skill?

The skill includes patterns for core packages like survival and survminer, as well as advanced libraries like flexsurv, rstpm2, tidycmprsk, and the tidymodels 'censored' package.

Does it support complex event analysis like competing risks?

Absolutely. It covers Cause-Specific Hazards and Fine-Gray Subdistribution Hazards using cmprsk and tidycmprsk for cases where multiple event types exist.

Can this skill help with plotting survival curves?

Yes, it provides comprehensive snippets for creating high-quality Kaplan-Meier plots, risk tables, and forest plots using the survminer and ggsurvplot frameworks.

R Survival Analysis Patterns

Name: R Survival Analysis Patterns
Author: choxos

bychoxos

データサイエンスとML

Performs comprehensive survival analysis and time-to-event modeling using modern R frameworks and tidy principles.

概要

This skill provides a robust framework for conducting survival analysis within the R ecosystem, offering standardized patterns for a wide range of statistical methods. It enables Claude to implement everything from basic Kaplan-Meier estimations and Cox proportional hazards models to advanced techniques like competing risks, multi-state models, and restricted mean survival time (RMST). With support for visualization libraries like survminer and integration with the tidymodels 'censored' package, this skill is essential for researchers and data scientists performing clinical, longitudinal, or reliability data analysis.

主な機能

Kaplan-Meier estimation with publication-quality survminer visualizations
Tidymodels integration for machine learning-based censored regression
Cox Proportional Hazards modeling with automated assumption testing
Flexible parametric models including Weibull and Royston-Parmar splines
Competing risks analysis using cause-specific and Fine-Gray methods
0 GitHub stars

ユースケース

Analyzing clinical trial data to determine treatment efficacy and patient survival rates
Predicting customer churn and time-to-attrition in subscription service models
Conducting industrial reliability engineering and equipment failure time analysis

概要

主な機能

Kaplan-Meier estimation with publication-quality survminer visualizations
Tidymodels integration for machine learning-based censored regression
Cox Proportional Hazards modeling with automated assumption testing
Flexible parametric models including Weibull and Royston-Parmar splines
Competing risks analysis using cause-specific and Fine-Gray methods
0 GitHub stars

ユースケース

Analyzing clinical trial data to determine treatment efficacy and patient survival rates
Predicting customer churn and time-to-attrition in subscription service models
Conducting industrial reliability engineering and equipment failure time analysis