Cox Proportional Hazards Calculator
Cox PH regression online with hazard ratios, Schoenfeld residuals and Kaplan Meier overlay.
When to use it
Use for time-to-event data with one or more covariates. Cox PH is the standard regression model in clinical and epidemiological survival analysis because it does not require specifying a baseline hazard.
When NOT to use it
Do not use when the proportional-hazards assumption fails (Schoenfeld test rejects); use a stratified Cox or time-varying coefficient. Do not use when there are very few events per covariate (rule of thumb at least 10 events per predictor; Peduzzi et al. 1995). For competing risks use a Fine-Gray model rather than cause-specific Cox.
What you get
Coefficient table with log-hazard estimates, SEs, Wald z, p, and 95 percent CIs; hazard ratios with their CIs; the model likelihood-ratio chi-square against the null; Harrell's concordance index (c-index); Schoenfeld residual test of proportional hazards globally and per covariate; Kaplan-Meier overlays.
How to interpret the output
A hazard ratio above 1 means the covariate increases the instantaneous risk; below 1 protective. Concordance index of 0.5 is no better than random; 0.7 plus is acceptable; 0.8 plus is strong (Harrell et al. 1996). A significant Schoenfeld test means the proportional-hazards assumption fails for that covariate; remedies include stratification or a time-varying coefficient. Report HRs with CIs, not p-values alone.
Common pitfalls
- Skipping the Schoenfeld PH check.
- Including too many predictors relative to the number of events (overfitting, unstable HRs).
- Forgetting that ties in event times require a tie-handling method (Efron is a good default).
- Treating Cox as a model for survival probabilities directly; it models the hazard, baseline survival is recovered separately.
Try it in StatVeda
The Cox Proportional Hazards (survival regression) engine runs entirely in the browser. No signup, no install, no data sent to a server. Paste your data, hit Run, copy the output.
Open Cox Proportional Hazards (survival regression)References
- Cox, D. R. (1972). Regression models and life-tables. Journal of the Royal Statistical Society B, 34(2), 187 to 220.
- Harrell, F. E. Jr. (2015). Regression Modeling Strategies, 2nd edition. Springer, New York.
- Therneau, T. M. and Grambsch, P. M. (2000). Modeling Survival Data: Extending the Cox Model. Springer, New York.