Parameters Estimation for the Unit log-log Distribution

Abstract

In this paper, point estimation problem of two unknown parameters of the unit log-log distribution is examined. For point estimation, six methods of estimate such as maximum likelihood, maximum product spacing, Anderson-Darling, least squares, weighted least squares, and Cramer-von Mises are examined in detail. Extensive simulation experiments are conducted to compare the effectiveness of these estimators based on bias and mean squared errors. According to the simulation results, it is seen that all estimators performed well in terms of two criteria and take close values in case of large sample. Moreover, practical data applications are performed for all estimators. Results of the Kolmogorov-Smirnov statistics are reported for all estimators in practical applications.