Generalized entropy optimization methods on lifetime data

In this thesis we have developed MinMaxEnt, MaxMaxEnt and MinMinxEnt, MaxMinxEnt (GEOM) analysis on survival data. Moreover, mentioned methods are compared with each other in the difference senses. It is observed that each of considered methods possessing an advantage over the other method in the given sense may be give up one's place to the same method in the another sense. The performances of improved methods are established by Chi – Square criteria, Root Mean Square Error (RMSE) criteria, Shannon entropy measure and Kullback-Leibler measure. According to obtained distributions (MaxMaxEnt)4, (MinMaxEnt)4,〖 (MinMinxEnt)4 and (MaxMinxEnt)4 estimator of Probability Density Function f(t), Cumulative Distribution Function F(t), Survival Function S(t) and Hazard Function h(t) are evaluated. The results are acquired by using statistical software MATLAB 7.10.0 (R2010a). The obtained results show that survival data analysis by GEOM acquires a new significance. Key words: survival function, censored observation, supplementary probability, MinMaxEnt, MaxMaxEnt, MinMinxEnt, MaxMinxEnt distributions