|
AUTHOR(S): M. Maswadah
|
TITLE Numerical Inference on the Generalized Gamma Distribution Parameters |
 PDF |
ABSTRACT In life data analysis, there are many methods for estimating distribution parameters. However, most of them are less efficient than Bayes’ method based on the informative prior. Therefore, the main objective of this study is to introduce optimal numerical iteration techniques, such as the Picard, Runge-Kutta and Adams methods for estimating the generalize life model parameters and compare them with the Bayes’ method based on the informative gamma and kernel priors. A comparison between these methods is provided using an extensive Monte Carlo simulation study based on two criteria, namely, the absolute bias and the mean squared error. The simulation results indicated that the numerical methods are highly favourable, which provide better estimates and outperform the Bayes’ method based on the generalized progressive hybrid censoring scheme. Finally, two real datasets analyses are presented to illustrate the efficiency of the proposed methods. |
KEYWORDS Bayes estimation; Adams’s method; Picard’s method; Runge-Kutta method; generalized progressive hybrid censoring scheme; informative prior; kernel prior |
|
Cite this paper M. Maswadah. (2025) Numerical Inference on the Generalized Gamma Distribution Parameters. International Journal of Mathematical and Computational Methods, 10, 260-276 |
|
