A0400
Title: A generalized closed-form maximum likelihood estimator for some survival distributions
Authors: Francisco Louzada - University of Sao Paulo (Brazil) [presenting]
Abstract: In modern applications, particularly those relying on embedded technology, real-time estimation and prediction are crucial. A novel generalization of the maximum likelihood estimator (MLE) is introduced, designed to provide closed-form solutions. This approach offers a significant advantage by enabling rapid computation without sacrificing accuracy. Numerical results demonstrate that the generalized MLE yields nearly unbiased estimates, even with small sample sizes, outperforming existing estimation procedures. The methodology is illustrated using Nakagami-type distributions, which are critical for modeling radio signal fading in communication engineering. Furthermore, the applicability of this method is extended to other important probability distributions, including the gamma and generalized gamma distributions, showcasing its broad utility in survival analysis and related fields.
