Abstract
In parameter estimation techniques the maximum likelihood estimation method is the most common technique used in social sciences and psychology although it is usually biased in a situation where sample sizes are small or when the data are heavily censored. Thus, the main objective of this paper is to present an optimal technique using the Runge-Kutta method to find the point estimation for the distribution parameters to avoid the drawback of the maximum likelihood estimation method. This method has been applied to derive the estimators of the inverse Weibull model parameters and compare them with the standard maximum likelihood estimation and Bayesian estimation methods based on the generalized progressive hybrid-censoring scheme, via the Monte Carlo simulations. The simulation results indicated that the estimates are highly favorable for the Runge-Kutta method, which provides better estimates and outperforms Bayesian and maximum likelihood estimation methods for different sample sizes and several values of the true parameters. Finally, two real data analyses are presented to demonstrate the efficiency of the proposed methods.
Author: Mohamed Maswadah
Accepted on: April, 2019
Accepted on: March, 2021