主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2023, Vol. 31 ›› Issue (12): 290-300.doi: 10.16381/j.cnki.issn1003-207x.2021.2707

• • 上一篇    

价格“同频共振”与产业链安全稳定

李坚飞1,唐昆1,沈炀2()   

  1. 1.湖南工商大学工商管理学院, 湖南 长沙 410205
    2.湖南工商大学经济与贸易学院, 湖南 长沙 410205
  • 收稿日期:2021-12-29 修回日期:2022-03-07 出版日期:2023-12-15 发布日期:2024-01-06
  • 通讯作者: 沈炀 E-mail:616553432@qq.com
  • 基金资助:
    国家社会科学基金资助项目(20BJY181);国家自然科学基金资助重大项目(71991463);湖南省自然科学基金资助项目(2021JJ30005)

Price Resonance and the Security and Stability of Hog Industry Chain: A Prediction Based on Price Mode Dissipation

Jian-fei LI1,Kun TANG1,Yang SHEN2()   

  1. 1.School of Business Administration, Hunan University of Technology and Business, Changsha 410205, China
    2.College of Economics and Trade, Hunan University of Technology and Business, Changsha 410205, China
  • Received:2021-12-29 Revised:2022-03-07 Online:2023-12-15 Published:2024-01-06
  • Contact: Yang SHEN E-mail:616553432@qq.com

摘要:

生猪产业链是一个典型耗散结构系统,产业链安全稳定与节点价格传导顺畅程度是该系统的内外熵流,决定了系统耗散结构的动态演化。本文基于耗散结构理论与隐马尔科夫模型,结合神经网络与机器学习中的前向算法、Baum-Welch算法以及Viterbi算法,提出了考虑产业链各节点价格模态耗散的生猪产业链稳定状态预测方法。通过STL-NAR组合模型预测生猪产业链中玉米、仔猪以及生猪价格,采用符号动力学将生猪产业链价格涨跌同频组合描述成不同价格模态,以确定产业链价格传导顺畅状态,并利用隐马尔科夫链HMM模型,评估和预测生猪产业链安全稳定状态及其异常风险。研究结果显示,STL-NAR模型对产业链节点价格传导顺畅程度具有良好的预测性能,能够为价格异常波动预测预警提供参考依据;生猪产业链节点价格“同频共振”与产业链安全稳定之间存在马尔科夫性质,系统内部产业链价格传导的顺畅程度能有效地预测生猪产业链的安全稳定,这为预测产业链安全稳定提供了科学研判的新思路。

关键词: 生猪产业链, 安全稳定, 耗散结构, 价格模态, 同频共振

Abstract:

The hog industry chain is a typical dissipative structure system. The security and stability of the chain is the internal entropy flow while the price resonance is external, which determines the dynamic evolution of the system. Based on dissipative structure theory and Hidden Markov model, combining neural network and forward algorithm, Baum-Welch algorithm and Viterbi algorithm in machine learning, a stable state prediction method of hog industry chain based on price modal dissipation of each node of the chain is proposed. The STL-NAR model is used to predict the prices of corn, piglet and hog. Then the price resonance of the chain is represented as different price modes to confirm the smoothness of price transmission through application of symbolic dynamics. The Hidden Markov Model is also applied to evaluate and predict the security and stability and abnormal risk of the chain. The empirical results show that, the STL-NAR model has good predictive performance for the smoothness of price transmission at each node of the industry chain, and it can provide a reference for the prediction and alertness of abnormal price fluctuations. There is a Markov property between the price resonance and the security and stability of the industry chain and the smoothness of price transmission can effectively predict the security and stability of the chain, which provides insights into future scientific research and judgment.

Key words: hog industry chain, security and stability, dissipative structure, price modes, resonance

中图分类号: