改进萤火虫算法在投资组合问题决策中的应用

Application of improved firefly algorithm in portfolio problem decision making

针对传统方法求解投资问题难度大、不易实现等问题,通过分析基本萤火虫算法缺点,设计了一种改进的自适应步长萤火虫算法用于求解该问题.利用正切函数使步长自适应变化,为算法加入了惯性线性因子,利用种群最优个体重新定义位置更新公式;为了避免算法陷入局部最优,为算法设计了两种变异操作;最后,将改进萤火虫算法应用于投资组合问题中.计算结果表明:改进萤火虫算法求解该问题具有明显优势.

Aiming at the difficulty of solving the investment problem by the traditional method,an improved adaptive step size firefly algorithm is designed to solve it by analyzing the shortcomings of the basic firefly algorithm.The tangent function is used to make the step size change adaptable.The inertial linear factor is added to the algorithm,and the position updating formula is redefined by using the optimal individual of the population.In order to avoid the algorithm falling into the local optimum,two kinds of mutation operations are designed for the algorithm.Finally,the improved firefly algorithm is applied to the portfolio problem,and the results show that the improved firefly algorithm has obvious advantages in solving the problem.