ON ESTIMATING PARAMETERS OF LINDLEY-GEOMETRIC DISTRIBUTION

Abstract

Lindley-geometric (LG) distribution is generated by compounding Lindley and geometric distribution. We tackle the problem of estimating parameters for LG distribution. For this purpose maximum likelihood, least-squares, weighted least-squares, Anderson-Darling and Crámer–von-Mises are used in order to estimate the two parameters of LG distribution. We also consider an extensive Monte Carlo simulation study to evaluate these methods according to the biases and mean-squared errors (MSEs). Finally, eight real data applications are presented.