Course contentsShow
← Data Science
Lesson 806 of 2,14521. Survival AnalysisFree lesson

Survival Functions and Hazard

Defining the survival function S(t) and hazard function h(t) as core concepts in survival analysis.

Survival Functions and Hazard

What you'll learn: You'll understand the two fundamental functions—survival and hazard—that form the mathematical foundation of survival analysis.

The Core Idea

In survival analysis, we need precise ways to describe when events happen over time. Two complementary functions give us this power:

The Survival Function S(t) answers: "What's the probability of surviving beyond time t?" It starts at 1.0 (everyone alive at the beginning) and decreases over time as events occur. Think of it like a countdown timer showing what proportion of subjects are still "event-free" at any given moment.

For example, if S(5 years) = 0.70 in a medical study, it means 70% of patients survived beyond 5 years.

The Hazard Function h(t) answers: "Given survival up to time t, what's the instantaneous risk of the event occurring right now?" Unlike the survival function which is cumulative, hazard captures the momentary risk at each time point. It's like checking your speedometer (instant speed) versus tracking total distance traveled.

If h(t) is high at a particular time, subjects face elevated immediate risk. If it's low, they're relatively safe at that moment.

The Relationship

These functions are mathematically linked—they describe the same underlying process from different angles. The survival function gives you the "big picture" probability, while the hazard function reveals the dynamic, time-varying risk pattern. Together, they help us understand not just if events happen, but when and how fast.

Key Takeaway: The survival function S(t) measures the probability of surviving beyond time t, while the hazard function h(t) captures the instantaneous risk of an event occurring at time t—both are essential tools for describing time-to-event data.