Analysis of Censored Life-tables with Covariates by Means of Log-linear Models

Abstract

In survival analysis the subject of observation is duration of time until some event called failure event. Often in such studies only partial information on the length of failure time is available what yields the so-called right-censored observations. The main interest in survival analysis is either to estimate the distribution of the true failure time or to identify the relationship between the true failure time and a set of some covariates. Additional troublesome point of theory and application of survival techniques is treatment of grouped observations (life-tables) along with incorporating covariates. In the paper a new approach is considered which allows to treat the censored life-table with qualitative covariates as a standard contingency table. Such a table can be further analysed by means of log-linear models or other standard multivariate inference techniques.

W pracy przedstawiono propozycję analizy tablicy trwania życia dla danych prawostronnie cenzurowanych. Przedstawiona metoda pozwala na sprowadzenie takiej tablicy do wielowymiarowej tablicy kontyngencyjnej, którą można analizować standardowymi technikami wielowymiarowego wnioskowania statystycznego, np. za pomocą modeli logarytmiczno-liniowych.