Parameter Estimation of Weibull Distribution Based on Second-Kind Statistics

Abstract

The log-cumulant estimator is proposed to estimate the parameters of Weibull distribution based on second-kind statistics. With the explicit closed form expressions, the log-cumulant estimator is computationally efficient. Parameterestimation results from Monte Carlo simulation and real synthetic aperture radar (SAR) image demonstrate that the log-cumulant estimator leads to better performance when compared to the moment estimator.

Keywords

• Weibull distribution
• parameter estimation
• second-kind statistics
• log-cumulant estimator