Modelling and testing of presence of hazard rates crossing under censoring

Mikhail Nikulin (1) , Vilijandas Bagdonavicius (2, 3) , Ruta Levuliene (4)

1 IMB - Institut de Mathématiques de Bordeaux
2 Departement of Statistics, Vilnius University
3 Departement of Statistics, Vilnius University
4 Departement of Statistics [Vilnius]

When analysing survival data from clinical trials, crossing of survival functions is sometimes observed .We present in our paper several models and estimation procedures for data with nonproportional hazards when data are censored.

Mots clés: Censored data; cross-effect of survivals; nonproportional hazard rates; Cox model

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Mikhail Nikulin, Vilijandas Bagdonavicius, Ruta Levuliene. Modelling and testing of presence of hazard rates crossing under censoring. Communications in Statistics - Theory and Methods, 2012, 41 (7), pp.980-991. ⟨hal-00783910⟩