TEI@I预测的有效性——来自持续五年公开预报珠三角港口运输需求项目的证据

管理评论 ›› 2020, Vol. 32 ›› Issue (7): 76-88.

TEI@I预测的有效性——来自持续五年公开预报珠三角港口运输需求项目的证据

田歆^1,2,3, 王皓晴^1,2,3, 朱佳仪^1,2,3, 鄂尔江⁴

1. 中国科学院大学经济与管理学院, 北京 100190;
2. 中国科学院虚拟经济与数据科学研究中心, 北京 100190;
3. 中国科学院大数据挖掘与知识管理重点实验室, 北京 100190;
4. 清华大学工业工程系, 北京 100084

收稿日期:2019-09-01 出版日期:2020-07-28 发布日期:2020-08-08
通讯作者: 田歆(通讯作者),中国科学院大学经济与管理学院副教授,中国科学院虚拟经济与数据科学研究中心副研究员,中国科学院大数据挖掘与知识管理重点实验室副研究员,硕士生导师,博士
作者简介:王皓晴,中国科学院大学经济与管理学院、中国科学院虚拟经济与数据科学研究中心、中国科学院大数据挖掘与知识管理重点实验室硕士研究生;朱佳仪,中国科学院大学经济与管理学院、中国科学院虚拟经济与数据科学研究中心、中国科学院大数据挖掘与知识管理重点实验室硕士研究生;鄂尔江,清华大学工业工程系博士研究生。
基金资助:
国家自然科学基金重大项目（71390331）；国家自然科学基金重点项目（71932002）；中国科学院大学优秀青年教师科研能力提升项目（Y8540XX162）。

The Effectiveness of TEI@I Forecasting: Evidence from a Five-year Project-based Public Prediction for Pearl River Delta Port Logistics

Tian Xin^1,2,3, Wang Haoqing^1,2,3, Zhu Jiayi^1,2,3, E Erjiang⁴

1. School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190;
2. Research Center on Fictitious Economy and Data Science, Chinese Academy of Sciences, Beijing 100190;
3. Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences, Beijing 100190;
4. Department of Industrial Engineering, Tsinghua University, Beijing 100084

Received:2019-09-01 Online:2020-07-28 Published:2020-08-08

摘要/Abstract

摘要： 突破传统采用历史数据训练来评估预测方法效度的局限，提出了一种基于真实情景中公开预报数据来评估预测有效性的框架。通过回顾分析一项长达五年公开发布的珠三角港口集装箱需求月度预报数据，检验了TEI@I预测方法论的有效性。实证结果表明，TEI@I预测方法论具有非常优良的预测精度和稳定性，预测效度随着预测时长的增加有所降低。同时论证了采用合理的预测方法，能够有效估计系统未来发展趋势。

关键词: TEI@I方法论, 预测, 港口物流, 集装箱吞吐量, 自适应性

Abstract: Existing literatures that assess the performance of forecasting methods typically based on historical data are partially subjective. We propose a brand new framework for evaluating and tracking the effectiveness of forecasting methods with data set from predicting project under real scenarios. This paper empirically examines the effectiveness of TEI@I methodology using public project-based data for Pearl River Delta Port logistics during the period from 2009 to 2013. Our analyses show that TEI@I model has both excellent prediction accuracy and robustness, which decreases with the increase of prediction duration. This paper also demonstrates that future development tendency of the complex system can be effectively predicted.

Key words: TEI@I methodology, forecasting, port logistics, container throughput, adaptivity

田歆, 王皓晴, 朱佳仪, 鄂尔江. TEI@I预测的有效性——来自持续五年公开预报珠三角港口运输需求项目的证据[J]. 管理评论, 2020, 32(7): 76-88.

Tian Xin, Wang Haoqing, Zhu Jiayi, E Erjiang. The Effectiveness of TEI@I Forecasting: Evidence from a Five-year Project-based Public Prediction for Pearl River Delta Port Logistics[J]. Management Review, 2020, 32(7): 76-88.

参考文献

[1] Wang S. Y. TEI@I:A New Methodology for Studying Complex Systems[R]. The International Workshop on Complexity Science, 2004
[2] Wang S. Y., Yu L., Lai K. K. Crude Oil Price Forecasting with TEI@I Methodology[J]. Journal of Systems Science and Complexity, 2005,18(2):145-166
[3] 余乐安,汪寿阳,黎建强. 外汇汇率与国际原油价格波动与预测——TEI@I方法论[M]. 长沙:湖南大学出版社, 2006
[4] 闫妍,许伟,部慧,等. 基于TEI@I方法论的房价预测方法[J]. 系统工程理论与实践, 2007,27(7):1-9
[5] 张嘉为,索丽娜,齐晓楠,等. 基于TEI@I方法论的通货膨胀问题分析与预测[J]. 系统工程理论与实践, 2010,30(12):2157-2164
[6] 柳冬,王雯珺,谢海滨,等. 我国房地产价格影响要素分析与趋势预测[J]. 管理评论, 2010,22(5):3-10
[7] 张茜,吴超,乔晗,等. 基于TEI@I方法论的中国季播电视综艺节目收视率预测[J]. 系统工程理论与实践, 2016,36(11):2905-2914
[8] Wang T., Tian X., Yu M., et al. Stage Division and Pattern Discovery of Complex Patient Care Processes[J]. Journal of Systems Science and Complexity, 2017,30(5):1136-1159
[9] 沈书立,李祥飞. TEI@I框架下的交通道路脆性预测模型研究[J]. 管理工程学报, 2018,32(2):240-247
[10] 田歆,曹志刚,骆家伟,等. 基于TEI@I方法论的香港集装箱吞吐量预测方法[J]. 运筹与管理, 2009,18(4):82-89
[11] Tian X., Lu X., Deng X. A TEI@I-Based Integrated Framework for Port Logistics Forecasting[C]. International Conference on Business Intelligence and Financial Engineering, 2009
[12] 田歆,汪寿阳,华国伟. 零售商供应链管理的一个系统框架与系统实现[J]. 系统工程理论与实践, 2009,29(10):45-52
[13] Granger C. W. J., Jeon Y. Long-term Forecasting and Evaluation[J]. International Journal of Forecasting, 2007,23(4):539-551
[14] Kahn H., Wiener A. J. The Year 2000——A Framework for Speculation on the Next Thirty-three year[M]. New York:MacMillan Company, 1967
[15] Tian X., Liu L. M., Lai K. K., et al. Analysis and Forecasting of Port Logistics Using TEI@I Methodology[J]. Transportation Planning and Technology, 2013,36(8):685-702
[16] Armstrong J. S. Combining Forecasts[M]. Boston:Springer, 2001
[17] Bates J. M., Granger C. W. J. The Combination of Forecasts[J]. Operations Research Quarterly, 1969,20(4):451-468
[18] Makridakis S., Spiliotis E., Assimakopoulos V. The M4 Competition:Results, Findings, Conclusion and Way Forward[J]. International Journal of Forecasting, 2018,34(4):802-808
[19] Aye G. C., Balcilar M., Gupta R., et al. Forecasting Aggregate Retail Sales:The Case of South Africa[J]. International Journal of Production Economics, 2015,160(1):66-79
[20] 柴建,张钟毓,李新,等. 中国航空燃油消费分析及预测[J]. 管理评论, 2016,28(1):11-21
[21] Makridakis S., Michèle H. The M3-Competition:Results, Conclusions and Implications[J]. International Journal of Forecasting, 2000,16(4):451-476
[22] Seabrooke W., Hui E. C. M., Lam W. H. K., et al. Forecasting Cargo Growth and Regional Role of the Port of Hong Kong[J]. Cities, 2003,20(1):51-64
[23] Xie G., Wang S. Y., Zhao Y., et al. Hybrid Approaches Based on LSSVR Model for Container Throughput Forecasting:A Comparative Study[J]. Applied Soft Computing, 2013,13(5):2232-2241
[24] Du P., Wang J. Z., Yang W. D., et al. Container Throughput Forecasting Using a Novel Hybrid Learning Method with Error Correction Strategy[J]. Knowledge-Based Systems, 2019,182(1):1-12
[25] Hyndman R. J., Athanasopoulos G. Forecasting:Principles and Practice[M]. Melbourne:OTexts, 2018
[26] Tian X., Liu L. M., Wang S. Y. Evolving Competition between Hong Kong and Shenzhen Ports[J]. Maritime Policy & Management, 2015,42(8):729-745