混沌理论和神经网络的农业经济预测仿真研究

Study on Farming Economic Forecasting Algorithm Based on Chaotic Theory and Neural Network

研究农业经济准确预测问题,农业经济具有混沌性、非线性的复杂系统,传统方法忽略农业经济变化的混沌性,难以精确地描述其变化规律,导致预测精度低。为提高农业经济时间序列预测的精度,提出混沌理论和神经网络相结合的农业经济预测模型(Chao-BPNN)。首先对农业经济变化的时间序列数据进行相空间重构,揭示隐含数据中的混沌变化规律,然后采用BP神经网络对农业经济时间序列数据进行学习和建模,反映农业经济变化的非线性变化点,并对神经网络参数进行优化,以提高预测精度。仿真结果表明,Chao-BPNN克服了传统方法的缺陷,能够全面、准确描述农业经济时间序列变化规律,提高了农业经济预测精度。

Study the problem of economic forecasting. Economic data are chaos, nonlinear time series, it is diffi- cult using a linear method to accurately describe the change rule. In order to improve economic forecasting precision, we proposed a nonlinear economic forecasting method based on the chaos theory. The economic time series data were analysed for chaos, phase space reconstruction, and revealing implicit data changes. Then the BP neural network was designed with economic time series data, and the neural network parameters were optimized to improve the accuracy of the prediction. Simulation experimental results show that this algorithm can overcome the defects of traditional methods, and is a comprehensive, accurate description of economic time series variation.