Optimization Method of Maximum Likelihood Eatimation Parameter Estimation Based on Fuzzy Genetic Algorithms
Sharina Huang · Modern Computer · 2009
Proposes a new approach hybrid of fuzzy genetic algorithm and maximum likelihood parameter estimation method for parameter estimation.The fuzzy logic is applied to tune the crossover and mutation probability of genetic algorithm,which is not be affected by the initial value,in this way we obtain the high solution precision and fast convergence speed.Finally parameter estimation is carried out based on Weibulls' three parameter distribution with GAOT toolbox.The result indicates that this method is better than traditional GA,the improved genetic algorithm has better balance between high efficiency and convergence to find global optimal solution,which can be better applied in mathematical statistics.