Statistical analysis of survival and reliability data with multiple crossing of survival functions

Vilijandas Bagdonavicius (1) , Mikhail Nikulin (2)

1 Departement of Statistics, Vilnius University
2 IMB - Institut de Mathématiques de Bordeaux

The so-called MCE model with possible multiple crossing of survival functions is studied. Under this model the ratios of the hazard rates increase, decrease or are constant, the hazard rates and the survival functions do not intersect, intersect once or twice .

Mots clés: Cox model; cross-effect of survivals; goodness-of-fit test; homogeneity; simple cross-effects model ratio of hazard rates; score statistic

Domaines: Statistiques [math.ST]; Théorie [stat.TH]; Applications [stat.AP]; Santé publique et épidémiologie

Vilijandas Bagdonavicius, Mikhail Nikulin. Statistical analysis of survival and reliability data with multiple crossing of survival functions. Comptes rendus de l'Académie des sciences. Série I, Mathématique, 2005, 340, pp.377-382. ⟨hal-00224124⟩