Estimation of the generalized lambda distribution from censored data

Abstract

Decision making based on censored data is a problem of serious concern as such data can provide only limited information compared to the corresponding uncensored data. In this article we describe a very general method of estimating the distribution of type I and II singly censored data. We derive single unified expressions for the probability weighted moments and L-moments of all univariate continuous distributions using a four-parameter family of distributions known as the Generalised Lambda Distribution (GLD) and use them to estimate the density of complete and censored data.