Estimation of the Shape Parameter for Lomax Distribution Under Bilateral Type-I Censored Sample

Abstract

The maximum likelihood estimation of shape parameters of Lomax distribution under bilateral Type-I Censored Sample is studied, the explicit expression of shape parameter estimation cannot be obtained, but it is proved that the maximum likelihood estimation of this parameter is unique. Then the EM estimation of shape parameters is obtained by using EM algorithm, and the asymptotic variance and approximate confidence interval of the EM estimation of shape parameters are also given; Finally, the effects of maximum likelihood estimation and EM estimation of shape parameters of Lomax distribution are numerically simulated. By comparing the relative deviation, it shows that the estimation effect of using EM algorithm to calculate shape parameters is relatively good, and the asymptotic variance and asymptotic confidence interval of EM estimation are given.