A MODELLING APPROACH TO INCREASE THE EXPLAINED RISK IN THE PROPORTIONAL HAZARDS REGRESSION

Authors : Deniz INAN, Oykum Esra ASKİN

Pages : 1-11

View : 39 | Download : 11

Publication Date : 2018-01-01

Article Type : Research Paper

Abstract :In this study, a modelling strategy is developed to obtain more information from censored obser- vations. By the proposed approach, uncensored observations are clustered using a fuzzy c-means algorithm and the degrees to which censored observations are members of these clusters are determined. Censored observations are weighted based on their membership values and the distances between the censoring time and the time components of the cluster centres. Further, simulation studies are performed to characterize the performance of the proposed approach based on the explained risk measure.
Keywords : Censored data, Explained risk, Fuzzy c means algorithm, Proportional hazards regression models, survival analysis