Definition
Survival analysis models time to an event — death, relapse, machine failure, customer churn — and can handle censored observations where the event did not occur during follow-up. The Kaplan–Meier estimate draws the survival curve, the log-rank test compares groups and Cox regression quantifies risk factors through hazard ratios.
Regression & models
Censoring is key: a patient who left the study healthy after two years carries the information "survived at least two years", and dropping them would bias the result. The survival curve shows the share of units without the event over time; median survival is the time at which it falls to 50 %.
The Cox proportional-hazards model does not require the shape of the time distribution, but it assumes the hazard ratio between groups is constant over time — verify that. Parametric models (Weibull, exponential) are used in product reliability and lifetime estimation.
In Statistica
Statistics → Advanced Linear/Nonlinear Models → Survival Analysis offers life tables, Kaplan–Meier curves with group comparisons (log-rank, Gehan, Cox–Mantel), Cox proportional-hazards regression and parametric models (exponential, Weibull, Gompertz). The Medical bundle complements it for clinical research.
Related terms
- Logistic regressionLogistic regression models the probability of a binary outcome — a patient responds to treatment or not, a customer…
- Generalized linear models (GLM)Generalized linear models extend linear regression to outcomes that are not normally distributed: binary (logistic…
- Nonparametric testsNonparametric tests do not assume a normal distribution — they work with the ranks of values rather than the values themselves.
Knowledgebase guides
FAQ
- What does a hazard ratio of 0.6 mean?
- That the risk of the event in the studied group is 40 % lower at every moment than in the reference group. A confidence interval containing 1 means a non-significant difference.
- Can I use survival analysis outside medicine?
- Yes — for time to equipment failure, customer churn or loan repayment; the principles of censoring apply the same way.
Try it on your own data
Statistica free for 30 days
Full version, no credit card. Or get a pricing estimate in a minute.
Regression & models
Updated: September 2026.