Home  /  Products  /  Features  /  Survival analysis
Order

Survival analysis

Analyze duration outcomes—outcomes measuring the time to an event such as failure or death—using Stata's specialized tools for survival analysis. Account for the complications inherent in this type of data such as sometimes not observing the event (censoring), individuals entering the study at differing times (delayed entry), and individuals who are not continuously observed throughout the study (gaps). You can estimate and plot the probability of survival over time. Or model survival as a function of covariates using Cox, Weibull, lognormal, and other regression models. Predict hazard ratios, mean survival time, and survival probabilities. Do you have groups of individuals in your study? Adjust for within-group correlation using a random-effects or shared-frailty model.

Cox proportional hazards

  • Time-varying covariates and censoring
  • Continuously time-varying covariates
  • Four ways to handle ties: Breslow, exact partial likelihood, exact marginal likelihood, and Efron
  • Robust, cluster–robust, bootstrap, and jackknife standard errors
  • Stratified estimation
  • Shared frailty models
  • Sampling weights and survey data
  • Multiple imputation
  • Martingale, efficient score, Cox–Snell, Schoenfeld, and deviance residuals
  • Likelihood displacement values, LMAX values, and DFBETA influence measures
  • Harrell’s C, Somers’ D, and Gönen and Heller’s K statistics measuring concordance
  • Tests for proportional hazards
  • Graphs of estimated survivor, failure, hazard, and cumulative hazard functions
  • Goodness-of-fit plot New

Cox proportional hazards model for interval-censored data Updated

Competing-risks regression

  • Fine and Gray proportional subhazards model
  • Time-varying covariates
  • Robust, cluster–robust, bootstrap, and jackknife standard errors
  • Multiple imputation
  • Efficient score and Schoenfeld residuals
  • DFBETA influence measures
  • Subhazard ratios
  • Cumulative subhazard and cumulative incidence graphs

Parametric survival models

  • Weibull, exponential, Gompertz, lognormal, loglogistic, or generalized gamma model
  • Robust, cluster–robust, bootstrap, and jackknife standard errors
  • Stratified models
  • Individual-level frailty
  • Group-level or shared frailty
  • Sampling weights and survey data
  • Multiple imputation
  • Martingale-like, score, Cox–Snell, and deviance residuals
  • Graphs of estimated survivor, failure, hazard, and cumulative hazard functions
  • Goodness-of-fit plot New
  • Predictions and estimates
    • Mean or median time to failure
    • Mean or median log time
    • Hazard
    • Hazard ratios
    • Survival probabilities

Interval-censored parametric survival models

  • Weibull, exponential, Gompertz, lognormal, loglogistic, or generalized gamma
  • Both proportional-hazards and accelerated failure-time metrics
  • Robust, cluster–robust, bootstrap, and jackknife standard errors
  • Stratified models
  • Sampling weights and survey data
  • Flexible modeling of ancillary parameters
  • Martingale-like, score, and Cox–Snell residuals
  • Graphs of estimated survivor, failure, hazard, and cumulative hazard functions
  • Goodness-of-fit plot New
  • Predictions and estimates
    • Mean or median time to failure
    • Mean or median log time
    • Hazard
    • Hazard ratios
    • Survival probabilities

Bayesian parametric survival models

  • Weibull, exponential, Gompertz, lognormal, loglogistic, or generalized gamma
  • Both proportional-hazards and accelerated failure-time metrics
  • Stratified models
  • Individual-level frailty
  • Group-level or shared frailty
  • Flexible modeling of ancillary parameters
  • Postestimation

Bayesian multilevel parametric survival models

  • Weibull, exponential, lognormal, loglogistic, or gamma
  • Both proportional-hazards and accelerated failure-time metrics
  • Two-, three-, and higher-level models
  • Nested and crossed random effects
  • Random intercepts and random coefficients
  • Flexible modeling of ancillary parameters
  • Postestimation

Finite mixtures of parametric survival models

  • Weibull, exponential, Gompertz, lognormal, loglogistic, or generalized gamma
  • Both proportional-hazards and accelerated failure-time metrics
  • Robust, cluster–robust, bootstrap, and jackknife standard errors
  • Sampling weights and survey data
  • Postestimation

Utilities

  • Create nested case–control datasets
  • Split and join time records
  • Convert snapshot data into time-span data

Features of survival models

  • Single- or multiple-failure data
  • Left-truncation
  • Right-censoring
  • Interval-censoring
  • Time-varying regressors
  • Gaps
  • Recurring events
  • Start–stop format
  • Different types of failure events
  • Customized time scales allowed

Random-effects parametric survival models

  • Weibull, exponential, lognormal, loglogistic, or gamma model
  • Robust, cluster–robust, bootstrap, and jackknife standard errors

Multilevel mixed-effects parametric survival models

  • Weibull, exponential, lognormal, loglogistic, or gamma models
  • Robust and cluster–robust standard errors
  • Sampling weights and survey data
  • Marginal predictions and marginal means

Treatment-effects estimation for observational survival-time data

  • Regression adjustment
  • Inverse-probability weighting (IPW)
  • Doubly robust methods
    • IPW with regression adjustment
    • Weighted regression adjustment
  • Weibull, exponential, gamma, or lognormal outcome model
  • Average treatment effects (ATEs)
  • ATEs on the treated (ATETs)
  • Potential-outcome means (POMs)
  • Robust, bootstrap, and jackknife standard errors

Structural equation models with survival outcomes

  • Latent predictors of survival outcomes
  • Path models, growth curve models, and more
  • Weibull, exponential, lognormal, loglogistic, or gamma models
  • Survival outcomes with other outcomes
  • Sampling weights and survey data
  • Marginal predictions and marginal means

Graphs of survivor, failure, hazard, or cumulative hazard function Updated

  • Kaplan–Meier survival or failure function
  • Nelson–Aalen cumulative hazard
  • Graphs and comparative graphs
  • Confidence bands
  • Embedded risk tables
  • Adjustments for confounders
  • Stratification
  • Interval-censored data Updated

Postestimation Selector

  • View and run all postestimation features for your command
  • Automatically updated as estimation commands are run

Life tables and analysis

  • Graphs and tables of estimates and confidence intervals
  • Mean survival times and confidence intervals
  • Cox regression adjustments
  • Actuarial adjustments
  • Tests of equality: log-rank, Cox, Wilcoxon–Breslow–Gehan, Tarone–Ware, Peto–Peto–Prentice, and Fleming–Harrington
  • Tests for trend
  • Stratified test

Power analysis

  • Solve for sample size, power, or effect size
  • Log-rank test of survival curves
  • Cox proportional hazards model
  • Clustered data
  • Exponential regression
  • See all power, precision, and sample-size features.

Obtain summary statistics, confidence intervals, etc.

  • Confidence intervals for incidence-rate ratio and difference
  • Confidence intervals for means and percentiles of survival time
  • Tabulate failure rate
  • Calculate person-time (person-years), incidence rates, and standardized mortality/morbidity ratios (SMR)
  • Calculate rate ratios with the Mantel–Haenszel or Mantel–Cox method

A survival example session

Additional resources

See New in Stata 18 to learn about what was added in Stata 18.