APSO_SVR模型在我国大豆价格预测的应用研究

发布时间：2018-04-30 07:05

本文选题：SVR预测模型 + 自适应　；参考：《大豆科学》2017年04期

【摘要】：国内大豆价格受到多重因素影响,如大豆进口量、国内大豆供给量、中国居民消费价格指数等,因此呈现非线性等特点。大豆价格的剧烈波动会对农户的种植结构和国家政策产生影响,因此准确预测大豆价格具有重要意义。支持向量回归(SVR)因其优越的寻优能力和较高的预测精确度而被广泛应用于非线性时间序列回归中。本文提出一种自适应粒子群算法(APSO)优化的SVR模型来预测我国大豆价格,该模型通过将现实空间内的数据映射到高维空间内,在高维空间内构造线性回归函数,从而判别原有空间内数据之间的关系。在SVR模型参数优化时,由于粒子群算法易陷入局部最优解,因此采用惯性权重更新和适应度变异的粒子群算法(APSO)对预测模型参数进行优化。采用2009年1月-2016年12月的国内大豆价格月度数据进行预测,结果表明APSO优化的SVR模型在大豆价格预测中精度较高,且能准确反应大豆价格的未来趋势,为从事大豆种植者及经营者提供决策依据。
[Abstract]:The domestic soybean price is affected by many factors, such as soybean import quantity, domestic soybean supply quantity, Chinese consumer price index and so on. The violent fluctuation of soybean price will have an impact on the farmers' planting structure and national policy, so it is of great significance to accurately predict the soybean price. Support vector regression (SVR) is widely used in nonlinear time series regression because of its superior optimization ability and high prediction accuracy. In this paper, an adaptive particle swarm optimization (SVR) optimization model is proposed to predict soybean prices in China. The model constructs a linear regression function in high-dimensional space by mapping the data in real space to high-dimensional space. In order to distinguish the relationship between the data in the original space. When the parameters of SVR model are optimized, the particle swarm optimization (PSO) algorithm is used to optimize the parameters of the prediction model because PSO is easy to fall into the local optimal solution, so the particle swarm optimization algorithm (PSO) which updates the inertia weight and the fitness mutation is used to optimize the parameters of the prediction model. The monthly data of soybean price in China from January 2009 to December 2016 are used to predict soybean price. The results show that the SVR model optimized by APSO has high precision in soybean price prediction and can accurately reflect the future trend of soybean price. To provide decision-making basis for soybean growers and operators.
【作者单位】：南京农业大学工学院;
【基金】：国家自然科学基金(71301077)
【分类号】：F323.7;TP18

【参考文献】