In survival data the response variate is the survival time of an individual like a medical patient or an industrial component. The responses are often *censored*, i.e. some individuals survive beyond the end of the study, and so their survival times are unknown. Genstat provides various ways of estimating the *survivor function* (i.e. the probability that an individual is still surviving at each time). You can do nonparametric tests to compare different survival distributions. Finally, you can model the survival times, by assuming that they follow exponential, Weibull or extremevalue distributions, or by fitting a proportional hazards model.

`KAPLANMEIER` |
calculates the Kaplan-Meier estimate of the survivor function |
---|---|

`RLIFETABLE` |
calculates the life-table estimate of the survivor function |

`RPHFIT` |
fits the proportional hazards model to survival data as a generalized linear model |

`RPHCHANGE` |
modifies a proportional hazards model fitted by `RPHFIT` |

`RPHDISPLAY` |
prints output for a proportional hazards model fitted by `RPHFIT` |

`RPHKEEP` |
saves information from a proportional hazards model fitted by `RPHFIT` |

`RPROPORTIONAL` |
fits a proportional hazards model by a direct maximization of the likelihood (this will be more efficient than `RPHFIT` for large data sets) |

`RSTEST` |
compares groups of right-censored survival data by nonparametric tests |

`RSURVIVAL` |
models survival times of exponential, Weibull or extreme-value distributions |