时态支持向量机模型在股票操纵模式发现上的研究

doi:10.12341/jssms22213

系统科学与数学 ›› 2023, Vol. 43 ›› Issue (2): 356-378.DOI: 10.12341/jssms22213

时态支持向量机模型在股票操纵模式发现上的研究

李博,孟志青,朱爱花

浙江工业大学管理学院, 杭州 310023

收稿日期:2022-03-30 修回日期:2022-08-24 出版日期:2023-02-25 发布日期:2023-03-16
通讯作者: 孟志青,Email:mengzhiqing@zjut.edu.cn
基金资助:
国家自然科学基金项目(11871434)资助课题.

PDF ( 1 ) 引用

李博,孟志青,朱爱花. 时态支持向量机模型在股票操纵模式发现上的研究[J]. 系统科学与数学, 2023, 43(2): 356-378.

LI Bo, MENG Zhiqing, ZHU Aihua. Research on Temporal Support Vector Machine Model in the Discovery of Stock Manipulation Patterns[J]. Journal of Systems Science and Mathematical Sciences, 2023, 43(2): 356-378.

Research on Temporal Support Vector Machine Model in the Discovery of Stock Manipulation Patterns

LI Bo, MENG Zhiqing, ZHU Aihua

School of Management, Zhejiang University of Technology, Hangzhou 310023

Received:2022-03-30 Revised:2022-08-24 Online:2023-02-25 Published:2023-03-16

基于原始时间属性下的时态数据难以发现规律的特点,文章构建了时态支持向量机模型,该模型通过对输入时态数据的粒度变换,获得多个分类模型,从而能够发现多种规律.在此基础上,结合时态型操纵特征构建了股票操纵模式发现模型,最后在证监会披露的操纵股票真实数据上进行数值实验,实验发现细时态粒度数据的分类模型在识别一般操纵和严重操纵上效果较好,粗时态粒度数据的分类模型在识别未被操纵或轻微操纵上效果较好.在未知数据集上实验,该模型可以有效识别不同程度操纵股票的模式,其中1个时态粒度数据下添加市场差异特征的模型表现最好,识别准确率达到了98.25%.文章验证了在不同时态粒度输入下,时态支持向量机模型能够发现在原始数据上不能发现的模式特征,这对解决一些复杂规律在原始特征下难以被发现的问题具有重要借鉴意义.

关键词： 股票操纵模式发现, 时态数据, 支持向量机, 操纵特征

Based on the fact that the temporal data under the original time attribute is difficult to find regularities, this paper constructs a temporal support vector machine model, which obtains multiple classification models by transforming the granularity of the input temporal data, so as to discover multiple regularities. On this basis, combined with the temporal manipulation features, a stock manipulation pattern discovery model is constructed. Finally, numerical experiments are carried out on the real data of manipulation stocks disclosed by the China Securities Regulatory Commission. The experiments show that the classification model of fine temporal granularity data is better in identifying general manipulation and serious manipulation, and the classification model of coarse temporal granularity data is better in identifying unmanipulated or slight manipulation. Experiments on unknown data sets show that the model in can effectively find the patterns of manipulation of stocks to different degrees. Among them, the model with the addition of market differences under one temporal granularity data performs the best, with a recognition accuracy rate of 98.25%. This paper verifies that under different temporal granularity inputs, the temporal support vector machine model can find pattern features that cannot be found in the original data, which has important reference significance for solving some complex laws that are difficult to find under the original features.

Key words: Stock manipulation pattern discovery, temporal data, support vector machine, manipulation feature

MR(2010)主题分类:

68Q32
62G08

分享此文:

[1] Eppen G D. Effect of Warehouse Centralization on Excepted Costs in a Multi-location Newsboy Problem. Management Science, 1979, 25(5):498-501.
[2] Chang P L, Lin C T. On the Effect of Centralization on Expected Costs in a Multi-location Newsboy Problem. Journal of the Operational Research Society, 1991, 42(11):1025-1030.
[3] Chen M S, Lin C T. Effects of Centralization on Expected Costs in a Multi-location Newsboy Problem. Journal of the Operational Research Society, 1989, 40(6):597-602.
[4] Cherikh M. On the Effect of Centralization on Expected Profits in a Multi-location Newsboy Problem. Journal of The Operational Research Society, 2000, 51(6):755-761.
[5] Kevin W Z. Risk-pooling Over Demand Uncertainty in the Presence of Product Modularity. International Journal of Production Economics, 1999, 62(1):75-85.
[6] Cai X, Du D.On the Effects of Risk Pooling in Supply Chain Management:Review and Extensions. Acta Mathematics Application Sinence, English Series, 2009, 25(4):709-722.
[7] Netessin S, Rudi N. Centralized and Competitive Inventory Models with Demand Substitution. Operations Research, 2003, 51(2):329-335.
[8] Schmitt A J, Snyder L V, Shen Z J M. Centralization Versus Decentralization:Risk Pooling, Risk Diversification, and Supply Chain Disruptions] [R]. Bethlehem:Working Paper, PC Rossin College of Engineering and Applied Sciences, Lehigh University, PA, 2011.
[9] Kemohliogluziya E, Bartholdi J. Centralizing Inventory in Supply Chains by Using Shapley Value to Allocate the Profits. Manufacturing Service Operations Management, 2011, 13(2):146-162.
[10] Swinney R. Inventory pooling with strategic consumers:Operational and behavioral] [M]. Stanford University Stanford CA Working Paper, 2012
[11] Yang Y, Chen Y, Zhou Y. Coordinating Inventory Control and Pricing Strategies Under Batch Ordering. Operations Research, 2014, 62(1):25-37.
[12] Zahra H, Babak A. The inventory centralization impacts on sustainability of the blood supply chain. Computers and Operations Research] 2018, 89:206-212.
[13] Shivangi V. T. Comparative study of Bullwhip effect in Centralized and Decentralized supply chain of an FMCG industry. Journal of Industrial Mechanics, 2018, 1(3):1-12.
[14] Christoph W, Achim K, Frank S, Leena S, Stefan V. A two-stage stochastic programming approach for identifying optimal postponement strategies in supply chains with uncertain demand. Omega, 2018, 83:123-138.
[15] Rockafellar R T, Uryasev S. Optimization of Conditional Value-at-Risk. The Journal of Risk, 2000, 2:21-41.
[16] Rockafellar R T, Uryasev S. Conditional value-at-risk for general loss distributions. Journal of Banking & Finance, 2002, 26:1443-1471.
[17] Alexander G J, Baptista A M. A Comparison of VaR and CVaR Constraints on Portfolio Selection with the Mean-Variance Model. Management Science, 2004, 50:1261-1273.
[18] Delage E, Ye Y. Distributional robust optimization under moment uncertainty with application to data-driven problems. Operations Research, 2010, 58(3):595-612.
[19] Huang D S, Zhu S S, Fabozzi F J, Fukushima M. Portfolio selection under distributional uncertainty:A relative robust CVaR approach. European Journal of Operational Research, 2010, 203:185-194.
[20] Zhu S S, Li D, Wang S Y. Robust Portfolio Selection under Downside Risk Measures. Quantitative Finance, 2009, 9:869-885.
[21] Zhu S S, Fukushima M. Worst-Case Conditional Value-at-Risk with Application to Robust Portfolio Management, Operations Research, 2009, 57(1):1155-1168.
[22] Takeda A, Kanamori T, A robust approach based on conditional value-at-risk measure to statistical learning problems. European Journal of Operational Research, 2009, 198(1):287-296.
[23] Gotoh J and Takano Y. Newsvendor solutions via conditional value-at-risk minimization. European Journal of Operational Research, 2007, 179(1):80-96.
[24] Zhou Y J, Chen X H, Wang Z R. Optimal ordering quantities for multi-products with stochastic demand:Return-CVaR model. International Journal of Production Economics, 2008, 112(1):782-795.
[25] Qiu R Z, Shang J, Huang X Y. Robust inventory decision under distribution uncertainty:A CVaR-based optimization approach. International Journal of Production Economics, 2014, 153(1):13-23.
[26] Sion M. On general minimax theorems. Pacific J. Mathematics, 1958, 8(1):171-176.
[27] Sawragi Y, Nakayama H, Tanino T. Theory of multiobjective optimization. Academic Press, New York, 1985.
[28] Meng Z Q, Dang C Y, Jiang M, Zhou G G, Shen R, Xu X S, Theory and Algorithm of Nonlinear Penalty Function. Science Press, Beijing, 2016.
[29] Xu L Y, Meng Z Q, Zhou G G, Mu Y Z, Zheng M C. Study on Single Cycle Production Allocation and Supply Strategy for DCEs Based on the CVaR Criterion. Discrete Dynamics in Nature and Society. 2018, 2018:1-15
[30] 徐蕾艳, 孟志青. 直营连锁企业生产分配供应单周期模型研究. 系统科学与数学, 2019, 39(6):904-917. (Xu L Y, Meng Z Q. Research on single cycle model of production allocation and supply for direct chain enterprises. Systems Science and Mathematics, 2019, 39(6):904-917.)
[31] 徐蕾艳, 孟志青. 基于凸概度分布簇下WCVaR模型的有限逼近性. 数学实践与认识, 2019, 49(4):245-251. (Xu L Y, Meng Z Q. Finite approximation of the worst conditional value-at-risk model based on convex probability distribution cluster. Mathematical in Practice and Theory, 2019, 49(4):245-251.)
[32] 徐蕾艳, 孟志青. 多损失WCVaR模型的等价性定理, 运筹学学报. 2018, 22(4):45-56. (Xu L Y, Meng Z Q. General equivalence problem of multi-loss WCVaR, Operations Research Transactions. 2018, 22(4):45-56.)

[1]	王冠鹏, 秦双燕, 崔恒建. 员工流失的影响因素分析与预测[J]. 系统科学与数学, 2022, 42(6): 1616-1632.
[2]	胡雪梅, 李佳丽, 蒋慧凤. 机器学习方法研究肝癌预测问题[J]. 系统科学与数学, 2022, 42(2): 417-433.
[3]	张婷婷, 王沫然, 魏得胜, 刘志峰. 季节调整FWA-SVR模型及其在旅游经济预测中的应用[J]. 系统科学与数学, 2021, 41(6): 1572-1584.
[4]	胡雪梅, 蒋慧凤. 具有技术指标的逻辑回归模型预测谷歌股票的涨跌趋势[J]. 系统科学与数学, 2021, 41(3): 802-823.
[5]	李萍，倪志伟，朱旭辉，宋娟. 基于改进萤火虫算法的SVR空气污染物浓度预测模型[J]. 系统科学与数学, 2020, 40(6): 1020-1036.
[6]	韩璐，苏治，刘志东. 金融市场的协动预测模型: DWT-SVM方法[J]. 系统科学与数学, 2020, 40(12): 2342-2356.
[7]	于静，韩鲁青. 一种改进的求解支持向量机模型的坐标梯度下降算法[J]. 系统科学与数学, 2018, 38(5): 583-590.
[8]	张文，崔杨波，姜祎盼. 基于SVM$^{K\text{-}{\rm Means}}$的非均衡P2P网贷平台风险预测研究[J]. 系统科学与数学, 2018, 38(3): 364-378.
[9]	李萍，倪志伟，朱旭辉，伍章俊. 基于分形流形学习的支持向量机空气污染指数预测模型[J]. 系统科学与数学, 2018, 38(11): 1296-1306.
[10]	唐振鹏，黄双双，陈尾虹. 基于支持向量机的银行系统重要性评估研究[J]. 系统科学与数学, 2018, 38(1): 57-77.
[11]	朱旭辉，倪志伟，倪丽萍，程美英，李敬明，金飞飞. 基于相异度的SVM选择性集成雾霾天气预测方法[J]. 系统科学与数学, 2017, 37(6): 1480-1493.
[12]	王勇，董恒新. 大数据背景下中国季度失业率的预测研究------基于网络搜索数据的分析[J]. 系统科学与数学, 2017, 37(2): 460-472.
[13]	张燕，张晨光，张夏欢. 平衡化图半监督学习方法[J]. 系统科学与数学, 2016, 36(8): 1107-1118.
[14]	朱旭辉，倪志伟，程美英. 基于人工鱼群和分形学习的雾霾天气预报方法[J]. 系统科学与数学, 2016, 36(11): 1887-1901.
[15]	张少白，曾又，刘友谊. 基于DIVA模型的脑电信号识别方法[J]. 系统科学与数学, 2015, 35(5): 489-498.

阅读次数

全文

摘要