经济管理学刊（中英文版）

基于马尔可夫预测模型的优化与实证分析

全文下载

作者：Xinping Yang, Wei Zheng, Yanmei Li, Yunyuan Yang

摘要：

The optimization model based on Markov chain is established to optimize the prediction of industrial structure and provide reference for policy adjustment. The vectorization operator is used to transform the Markov prediction model into an optimization problem with constraints, which highlights the theoretical proof and computational rigor. Based on the data of three industrial structures in Yunnan Province from 1989 to 2019, this paper establishes Markov optimization model to predict the proportion of three industrial structures in Yunnan Province from 2020 to 2030. The maximum percentage average absolute error and hill inequality coefficient of the prediction are 1.2335% and 0.2, respectively. The order-degree of the three industrial structures is a stable series, which is stable around 1 after 1996. The sample data and the predicted values show four stages of change characteristics. After 2020, the three industrial structures are stable in the "three, two and one" structure.

关键词：Markov Chain; Constrained Optimization; Three Industrial Structures in Yunnan Province; Degree of Order; Prediction

参考文献：

[1]      Peng, JZ, Sun, ZM, and Huang, X.2015, “Empirical analysis of industry structure and economy’s synergetic development based on grey relational theory: evidence from Jiangxi Province”, Economic Geography, vol.35, no.8, pp.123-128.

[2]      Wang, L. “The economic structure in Hunan Province based on the gray system theory”, master thesis, Changsha: Central South University, 2013.

[3]      Zhou, QW, Li, HM, and Jia, YF.2007,“ Application of the grey system in the restructuring of regional industrial restructuring”, Journal of Xian University of Architecture & Technology(Natural Science Edition),vol.39,no.3,pp.437-439．

[4]      Zhao, D. “Empirical study of industrial structure upgrading and economic growth in western underdeveloped areas based on Markov theory”, Mathematics in Practice and Theory, vol.49, no.22, pp.9-14.

[5]      Zhao, D, and Long, Q. 2015, “Investigation on forestry industrial structure in Guizhou Province based on Markova optimization model”, Issues of forestry economics, vol.35, no.4, pp.318-322.

[6]      Li, J. and Sheng, XZ. 2018, “Research on industrial structure of Yunnan Province based on Markov chain”, Rural Economy and Science-technology, vol.29, no.4, pp.177.

[7]      Hao, LL. Li, LH. Zhang, B, and Song, H. 2019, “Cluster analysis of regional industrial structure change in Heilongjiang Province based on Markov transfer matrix”, Journal of Northeast Normal University (Natural Science Edition), vol.51, no.4, pp.131-136.

[8]      Wang, R, Luo, HL, and Xie, CM. 2015, “Panzhihua's industrial structure based on shift-share method”, Resources &Industries, vol.17, no.6, pp.132-137.

[9]      Bao, YW, and Xiong, HH. 2015, “Analysis of urban industrial structure based on dynamic shift share analysis: a case study of Shiyan City”, Journal of China Three Gorges University Humanities and Social Science, vol.37, no.2, pp.65-70．

[10]   Xu, DY, and Yang, JZ. 2013, “On forestry industry structure of Fujian Province based on Shift-share method”, Journal of Fujian Agriculture and Forestry University (Philosophy and Social Science Edition), vol.16, no.2, pp.41-44.

[11]   Xie, YM, Zhang, J, and Zhang, QQ.2012,“Shift-share analysis of forestry industry structure in Yunnan Province”, Journal of Southwest Forestry University, vol.32, no.3, pp.75-79. 

[12]   Li, DF. Statistical computing, Beijing: Higher Education Press, 2017, pp.243-249.

[13]   Gao, L. Numerical optimization method, Beijing: Peking University Press, 2014, pp.217-226.

[14]   Berwin A., Turlach R., and Andreas Weingessel.2007, quadprog: Functions to solve quadratic programming problems. R package version 1.4-11.

[15]   Dang, YG, Liu, SF, and Wang, QF. Theory and practice of regional industrial structure optimization, Beijing: Science Press,2011.

[16]   Yi, DH. Data analysis and Eviews application (Second Edition), Beijing: China Renmin University Press, 2014, pp.39-45; 145-160.