Kernel Inference on the Weibull Distribution Parameters and Reliability

Abstract A new estimation technique based on non-parametric kernel density estimation is introduced as an alternative and reliable technique for estimation in life testing models. This technique estimates the parameters and reliability directly from the data without making any prior assumptions. The efficiency of this technique has been studied in comparison to Bayesian estimation of the parameters and the reliability of the Weibull distribution based on informative and informative conjugate priors, via Monte Carlo simulations. The simulation results indicated the robustness of the proposed method over Bayes’ method. Finally, a numerical example is given to illustrate the densities and the inferential methods developed in this paper.