Modeling engineering data using extended power-Lindley distribution: Properties and estimation methods
Loading...
Date
2021
Journal Title
Journal ISSN
Volume Title
Publisher
Journal of King Saud University - Science
Abstract
In this paper, we introduce a new flexible distribution called the Weibull Marshall-Olkin power-Lindley
(WMOPL) distribution to extend and increase the flexibility of the power-Lindley distribution to model
engineering related data. The WMOPL has the ability to model lifetime data with decreasing, increasing,
J-shaped, reversed-J shaped, unimodal, bathtub, and modified bathtub shaped hazard rates. Various properties
of the WMOPL distribution are derived. Seven frequentist estimation methods are considered to
estimate the WMOPL parameters. To evaluate the performance of the proposed methods and provide a
guideline for engineers and practitioners to choose the best estimation method, a detailed simulation
study is carried out. The performance of the estimators have been ranked based on partial and overall
ranks. The performance and flexibility of the introduced distribution are studied using one real data
set from the field of engineering. The data show that the WMOPL model performs better than some
well-known extensions of the power-Lindley and Lindley distributions.
Description
Keywords
Anderson–Darling estimation Maximum likelihood estimation Maximum product of spacing Moments Power-Lindley distribution