大数据时代的数量经济模型研究——以BP神经网络的中国CPI预测为例

Research on Quantitative Economic Model in the Era o f Big Data--Taking the prediction of China's CPI based on BP neural network as an example

大数据时代的来临,掀起了大数据与传统学科融合发展的热潮,探寻"大数据+数量经济学模型"的发展之路势在必行。本文运用文献计量分析,对大数据时代较有代表性的定量研究方法和模型进行分类梳理,并基于BP神经网络模型结合36个指标数据对中国居民消费价格指数(CPI)进行多元变量预测。研究表明,BP神经网络的预测结果最优。因此,未来要强化传统数量经济学的大数据化升级,注重非线性、非参数、模型自由化的研究方法在实际问题中的应用。

The advent of the era of big data has set off an upsurge of the integration and development of big data and traditional disciplines. It is imperative to explore the development road of "big data + quantitative economics model". In this paper, bibliometric analysis is used to classify and sort out the representative quantitative research methods and models in the era of big data. Based on BP neural network model, combined with 36 index data, multivariate prediction of China’s consumer price index(CPI) is carried out. The results show that the BP neural network has the best prediction results, and its prediction accuracy is even much higher than the average CPI forecast published by wind. Therefore, in the future, we should strengthen the big data upgrading of traditional quantitative economics, and pay attention to the application of nonlinear, nonparametric and model liberalization research methods in practical problems.