Proposal of Using Principle of Maximizing Entropy of Generalized Gamma Distribution to Estimate the Survival probabilities of the

Abstract

In this research we been estimated the survival function for data suffer from the disturbances and confusion of Iraq Household Socio-Economic Survey: IHSES II 2012 , to data from a five-year age groups follow the distribution of the Generalized Gamma: GG. It had been used two methods for the purposes of estimating and fitting which is the way the Principle of Maximizing Entropy: POME, and method of booting to nonparametric smoothing function for Kernel, to overcome the mathematical problems plaguing integrals contained in this distribution in particular of the integration of the incomplete gamma function, along with the use of traditional way in which is the Maximum Likelihood: ML. Where the comparison on the basis of the method of the Central Bureau of Statistics to stay through the program MORTPAK real function values calculated. And then compared to the use of Root Mean Square Error: RMSE, and Mean Absolute Percent Error: MAPE. The results showed preference entropy as optimal method to estimate survival function on other methods.