Abstract:

A non-Bayesian approach to parameter estimation, statistical inference and decision-making are discussed and compared. A pragmatic criterion, success in practice, as well as logical consistency is emphasized in comparing alternative approaches. In this study, attention is given to skew distribution for modelling lifetime data in particular, Marshal Olkin Alpha Power Inverse Exponential (MOAPIE) distribution. Parameters of the distribution were estimated using non-Bayesian estimation methods of Maximum Likelihood Estimation, Least Square Estimation and Weighted Least Square Estimation. Finally, simulated and real life data applications illustrate the performance of the estimation methods. Keywords: Non-Bayesian estimation, Maximum Likelihood estimation, Least Square estimation, Weighted Least Square estimation, Simulation.