BAYESIAN WEIBULL MIXTURE MODELS FOR DENGUE FEVER
Sri Astuti Thamrin, Jaya, Andi Kresna, La Podje Talangko · Hasanuddin University Repository · 2014
Dengue fever disease has become common problem in developing countries including Indonesia. Mixture models are usually used in modelling data consisting of several groups, where each group has different properties and characteristics of the one family but uses the same distribution. Weibull mixture models have received increasing attention in recent statistical research with applications in the field of survival analysis. The advances in the Bayesian paradigm have substantially expanded the methodology and application of Weibull mixture models. One problem of current interests is the analysis of survival times of patients. Dengue hemorrhagic fever data can be used to make inference about patient survival. In this study, we focus on the use of Bayesian Weibull mixture models for estimating survival. A simulation study that investigates the impact of censoring on these models is also described.