Estimation of the coefficients of variation for inverse power Lomax distribution

Abstract

One useful descriptive metric for measuring variability in applied statistics is the coefficient of variation (CV) of a distribution. However, it is uncommon to report conclusions about the CV of non-normal distributions. This study develops a method for estimating the CV for the inverse power Lomax (IPL) distribution using adaptive Type-Ⅱ progressive censored data. The experiment is a well-liked plan for gathering data, particularly for a very dependable product. The point and interval estimate of CV are formulated under the classical approach (maximum likelihood and bootstrap) and the Bayesian approach with respect to the symmetric loss function. For the unknown parameters, the joint prior density is calculated using the Bayesian technique as a product of three independent gamma densities. Additionally, it is recommended to use the Markov Chain Monte Carlo (MCMC) method to calculate the Bayes estimate and generate posterior distributions. A simulation study and a numerical example are given to assess the performance of the maximum likelihood and Bayes estimations.

Keywords

• gibbs and metropolis sampler
• inverse power lomax distribution
• adaptive type-ⅱ progressive censoring scheme
• coefficient of variation
• bayesian approach