混沌算子网络在经济数据预测中的应用研究

Economic data forecast based on chaotic operators network

采用多个混沌算子单元组成一种新的预测网络,实现经济数据的预测分析。利用已知数据构造出训练样本,通过调节混沌算子单元的控制参数来控制其动力学特性,以此改变预测网络的动力学行为,使预测网络的动力学特性逐渐逼近被预测系统,并随之一致变化,从而实现时间序列的动态预测分析。利用该方法对国内生产总值和国民总收入等经济数据进行了预测研究,结果表明,与传统的静态预测模型相比,该方法具有更好的预测性能和预测效果,能够有效实现经济数据的短期预测。预计未来几年我国国内生产总值和国民总收入将保持持续增长的趋势。

Forecast analysis of economic data is realized by a new prediction network composed of multiple chaotic operators.Training samples are constructed by given data,and the dynamic characteristics of chaotic operator is controlled by adjusting control parameters.In this way,the dynamic behavior of the prediction network is changed to approach to the predicted system.Thereby,dynamic forecasts of some economic data,such as gross domestic product,gross national income,are realized by the prediction model.The experiment results show that the presented method can realize short-term forecast of economic data effectively,and it has better predictive performance and effect,compared to traditional static prediction methods.In the coming years,it is predicted thal gross domestic product and gross national income will maintain the sustained growth.