Log-Cumulant Parameter Estimator of Log-Normal Distribution

摘要

Based on the second-kind statistics, the log-cumulant estimator is proposed for parameter estimation of the log-normal distribution. The estimates of the shape parameter and scale parameter are independent for the log-cumulant estimator. Parameter estimation results from Monte Carlo simulation demonstrate that, the log-cumulant estimator is insensitive to samples and it leads to better performance when compared to the moment estimator.