An Alternative Estimator with Appropriate Plotting Position Estimates for the Generalized Exponential Distribution

Abstract

In this paper, we propose an alternative method for estimating generalized exponential distributions by applying plotting positions of modified percentile estimates. We compared its efficiency with the classical maximum likelihood estimator and percentile estimator in terms of root mean square errors. Simulation results show that the percentile estimator outperformed the others for small sample sizes while the proposed estimator was most effective for medium to large sample sizes. This finding was supported by applying the proposed estimator to deduce whether Thai rainy season rainfall data followed a generalized exponential distribution from a rain gauging station in Fang district, Chiang Mai Province, Thailand.