基于改进PSO的组合预测模型研究

为进一步提高组合预测的预测精度，有必要对预测模型的权重分配进行研究。将粒子群算法用于求解组合预测中模型的权重，并在研究过程中针对基本粒子群算法的不足，对粒子群算法的参数惯性权重和加速度因子进行了改进，构造了基于改进粒子群算法的组合预测模型。以重庆市物流需求的预测为背景，以四种方法为参照对象，对比验证了该改进模型的有效性以及预测的准确性。

Research of combined forecasting model based on improved PSO

In order to improve the accuracy of combined forecasting, it is necessary to study the weights allocation of forecasting model. The paper uses Particle Swarm Optimization （PSO） to solve the weight of combination forecast model. In the process of research, aiming at the shortcomings of the basic particle swarm algorithm, inertia weight and acceleration factor of the parameters of particle swarm optimization are improved. And the combination forecast model based on improved particle swarm algorithm is constructed. To Chongqing logistics demand forecasting as the background, with four methods as reference object, it compares to verify the effectiveness of the improved model and the accuracy of prediction.