Maximum Likelihood Parameter Estimation for the Exponentially- Modified Logistic Distribution Based on Particle Swarm Optimization

Abstract: Exponentially-modified logistic distribution is a new flexible modified distribution. It is regarded as a strong competitor for widely used classical symmetrical and non-symmetrical distributions such as normal, logistic, lognormal, and log-logistic. In this study, the unknown parameters of the distribution have been estimated using the maximum likelihood method. Meta-heuristic algorithms have been used to solve the nonlinear equations of this method. The algorithms used in this study are the Sine Cosine and the Particle Swarm Optimization Algorithms. The efficiencies of maximum likelihood estimates for these algorithms are compared via a Monte-Carlo simulation study. It has been seen that the likelihood estimates for the location α and scale β parameters of the exponentiallymodified logistic distribution developed with the Particle Swarm Optimization algorithm are more efficient than the Sine Cosine algorithm.