Parameter estimation for the 3-parameter Gamma distribution usingthe continuation method

摘要

This paper shows a maximum-likelihood (ML) parameter estimationnalgorithm for the 3-parameter Gamma distribution. The algorithm, ancombination of the continuation method and the extended Gammandistribution model, can find the local ML estimates of the parametersnwithout a careful selection of the starting point in the iterativenprocess. This algorithm is more efficient than previous algorithms, andncan find the multiple local ML estimates