Bayesian frailty model for time to event breast cancer data.

Leo Alexander, T.; Pari Dayal, L.; Valarmathi, S.; Ponnuraja, C.; Venkatesan, P.

Indian Journal of Applied Research; 2014; 4; 8-12.


Abstract: Survival analysis has become standard tools for modeling cancer trial data when the event of interest is the “time to event”. In survival analysis, the proportional hazard model was introduced by Cox (1972) in order to estimate the effects of different covariates influencing the time-to-event data. This model has been used extensively in time to event of cancer trial data. It is known that the Bayesian analysis has the advantage in dealing with small sample of censored data over frequentist methods. Frailty models in survival analysis deal with the unobserved heterogeneity among subjects. The objective of this article is to present a Bayesian analysis for survival models with frailty and is being compared with a frequentist method of proportional hazards model. Gibbs sampling technique is used to assess the posterior quanti­ties of interest and to avoid the complexity in calculations. The posterior is arrived using WinBUGS package. An illustrative analysis is done within the context of survival time to death of breast cancer data.


Keywords: Survival data; frailty model; Bayesian approach; Gibbs sampler; WinBUGS




Back to List of publications / Home