Cox Proportional Hazards Model

The Cox proportional hazards model is a semiparametric regression model for time-to-event data that estimates how predictors multiply the hazard of an event without specifying the baseline hazard.

콕스 비례위험모형은 기저위험을 지정하지 않고 예측변수가 사건 위험을 몇 배로 변화시키는지를 추정하는 사건발생시간 자료의 준모수 회귀모형이다.

How do I find which factors affect the time until an event while handling censored cases?

중도절단된 사례를 고려하면서 사건이 일어나기까지의 시간에 영향을 주는 요인을 어떻게 찾나요?

The Cox proportional hazards model relates predictors to the time until an event, such as death or churn, while correctly using censored observations. It reports hazard ratios and fits when each predictor's effect on the hazard is roughly constant over time.

When to use Cox Proportional Hazards Model

• Time-to-event outcome with censoring, such as survival, relapse or customer churn
• Several continuous or categorical predictors of risk
• Reporting adjusted hazard ratios

When to use something else

• Hazards that clearly cross or effects that change over time; use time-varying coefficients or accelerated failure time models
• Competing event types that preclude the event; use competing risks regression

Assumptions and requirements

• Proportional hazards (checked with Schoenfeld residuals)
• Non-informative censoring
• Independent observations
• Linear effect of continuous predictors on the log hazard

What the analysis reports

• Hazard ratios with 95% confidence intervals
• Wald, likelihood ratio and score tests
• Concordance index
• Schoenfeld residual test of proportional hazards
• Adjusted survival curves

How to report it

In the Cox model, [predictor] was associated with a higher hazard of [event], HR = [value], 95% CI [[value], [value]], p = [value].

Related methods

• Kaplan-Meier Estimator
• Log Rank Test
• Accelerated Failure Time Model

Run Cox Proportional Hazards Model on Quorant