Survival Data Analysis By Generalized Entropy Optimization Methods

Abstract

Entropy Optimization Methods (EOM) have important applications, especially in statistics, economy, engineering and etc. There are several examples in the literature that known statistical distributions do not conform to statistical data, however the entropy optimization distributions do conform well. It is known that all statistical distributions can be obtained as the MaxEnt distribution and Entropy Optimization Distribution (EOD) especially as Generalized Entropy Optimization Distribution (GEOD) more exactly represents the given statistical data. In this paper, survival data analysis is fulfilled by applying Generalized Entropy Optimization Methods (GEOM). GEOM have suggested distributions in the form of the MinMaxEnt, the MaxMaxEnt which are closest and furthest to statistical data in the sense of information theory respectively. In this research, the data of male patients with localized cancer of a rectum diagnosed in Connecticut from 1935 to 1944 is considered and the results are acquired by using statistical software R and MATLAB. The performances of GEOD are established by Chi - Square, Root Mean Square Error (RMSE) and Information criteria.