Red Door Analytics
Resources · Primer

What is the Cox model?

An introduction to the Cox proportional hazards model, the most widely used statistical tool for analysing survival data.

Primer2 min read

The Cox model, also known as the proportional hazards model, is a popular statistical tool used to analyse survival data. It was developed by British statistician Sir David Cox, and published in 1972.

hazard h(t) time ×1.7 h₀(t) h₀(t)e βx the ratio is the same at every time the baseline shape is left unspecified
The Cox model leaves the baseline hazard h₀(t) completely unspecified — drawn here as an arbitrary dashed shape — and estimates only how much a covariate multiplies it. Both curves share that shape; only the vertical scaling differs.

It has gained popularity largely by avoiding making parametric assumptions about the shape of the baseline rate in a survival model.

The Cox model estimates the relationship between a set of covariates and the hazard function: the instantaneous rate at which the event occurs at a given time, among those who have not yet had it. It is a rate, not a probability, and can be larger than one.

The Cox model is a semi-parametric model, meaning that it makes some assumptions about the underlying distribution of the data but does not require complete specification of the distribution.

Instead, it assumes that the hazard function is a product of two components: a baseline hazard function that describes the underlying risk of the event over time, and a set of covariate effects that modify the baseline hazard function for each individual.

Modelling is done on the scale of the hazard function, h(t), which is defined as:

h(t)=h0(t)exp(Xβ)

where h0(t) is the baseline hazard function, and Xβ the linear predictor representing the covariate effects.

A regular Cox model assumes proportional hazards, which means that the ratio of the hazard functions for any two individuals is constant over time. However, this assumption can be relaxed.

Need this applied to your own data?

Tell us what you're modelling and we'll point you to the closest worked example — or build one with you.

Get in touch Start a project