Evaluation and Comparison of Metaheuristic Methods to Estimate the Parameters of Gamma Distribution

Parameter estimation of three parameter (3-p) Gamma distribution is very important as it one the most popular distributions used to model skewed data. Maximum Likelihood (ML) method based on finding estimators that maximize likelihood function, a well-known method. It rather difficult function formed for 3-p distribution. In this study, five well known metaheuristic methods, Simulated Annealing (SA), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Differential Evolution (DE), and Artificial Bee Colony (ABC), are suggested obtain ML estimates parameters Monte-Carlo simulations performed examine efficiencies methods problem Also, differences between solution qualities computation time algorithms investigated by statistical tests. Moreover, multi-criteria decision-making Technique Order Performance Similarity Ideal Solution (TOPSIS), preferred ranking according their performance in estimation. Results show superior others consideration all criteria quality, time, simplicity, robustness algorithms. addition, an analysis real-life data presented demonstrate implementation methods.