Widening valid estimation range of multilook Pareto shape parameter with closed-form estimators

Two improved estimators for estimating the shape parameter of multilook Pareto distribution are proposed. The proposed estimators are aimed at widening the valid estimation range of the shape parameter by using the assemble averages <z^r> and <z^r log z>. Moreover, the estimators are closed-form if r takes the negative integers. Compared with the existing estimators which are either non-closed-form or invalid for small shape parameters, the proposed estimators can estimate the shape parameter in the whole range from zero to infinity with closed forms. Simulation results demonstrate the better estimation performance of the proposed estimators.