Kernel Inference on the Generalized Gamma Distribution Based on Generalized Order Statistics

• Main Authors: M. Ahsanullah, M. Maswadah, Ali M. Seham
• Format: Article
• Language: English
• Published: Atlantis Press 2013-08-01
• Series: Journal of Statistical Theory and Applications (JSTA)
• Subjects: Generalized gamma distribution; Generalized order statistics; Maximum likelihood estimation; Kernel density estimation; Asymptotic maximum likelihood estimations.
• Online Access: https://www.atlantis-press.com/article/8356.pdf

The kernel approach has been applied using the adaptive kernel density estimation, to inference on the generalized gamma distribution parameters, based on the generalized order statistics (GOS). For measuring the performance of this approach comparing to the Asymptotic Maximum likelihood estimation, the confidence intervals of the unknown parameters have been studied, via Monte Carlo simulations, based on their covering rates, standard errors and the average lengths. The simulation results indicated that the confidence intervals based on the kernel approach compete and outperform the classical ones. Finally, a numerical example is given to illustrate the proposed approaches developed in this paper.