Please wait a minute...
Advanced Search
数据分析与知识发现  2019, Vol. 3 Issue (10): 118-126    DOI: 10.11925/infotech.2096-3467.2019.0192
  研究论文 本期目录 | 过刊浏览 | 高级检索 |
基于股票日内交易加权网络的超短期股票交易型操纵识别研究 *
扈文秀,马丽,张建锋()
西安理工大学经济与管理学院 西安 710054
Identifying Ultra-short-term Market Manipulation with Stock Intraday Trading Weighted Network
Wenxiu Hu,Li Ma,Jianfeng Zhang()
School of Economics and Management, Xi’an University of Technology, Xi’an 710054, China
全文: PDF(631 KB)   HTML ( 11
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】构建股票日内交易加权网络, 选取刻画网络特征的主要参数, 识别超短期股票交易型操纵。【方法】基于股票日内Tick交易数据, 以买卖双方的委托报单ID为节点, 以委托报单是否成交为连线, 以双方的实际成交量为权值构建股票日内交易加权网络。采用Pajek5.03、Ucinet6两款分析软件获得复杂网络统计参数, 进而构建股票超短期交易型操纵识别模型。【结果】实证分析结果表明, 加权平均节点度、网络密度等9项网络参数是判断公司股票是否被实施超短期交易型操纵的主要识别参数, 识别模型样本内与样本外检验整体准确率分别为93.58%与87.73%。【局限】仅选取2015年牛市的样本, 未收集到熊市样本共同分析。【结论】本文所构建模型解决了超短期股票交易型操纵难以识别的问题, 为证券监管部门准确打击市场操纵行为提供技术支持。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
扈文秀
马丽
张建锋
关键词 日内交易加权网络逻辑回归股票交易型操纵识别模型    
Abstract

[Objective] This paper proposes a weighted network for stock intraday trading, and selects the main parameters for network features, aiming to identify the ultra-short-term market manipulations. [Methods] We constructed the weighted network for stock intraday trading with tick data, and used the order ID as nodes. The lines of network were the dealing orders, and the weights of line values were actual trading volumes. Analytical software Pajek5.03 and Ucinet6 were used to obtain the statistical parameters of complex networks for the proposed model. [Results] The nine network parameters, such as weighted average degree and network density, can be used as the main parameter to determine the stock manipulation. The overall accuracy values of our model with internal and external samples were 93.58% and 87.73%. [Limitations] We only retrieved the bull market data from 2015, while the bear market data were not collected. [Conclusions] This study helps authorities identify and crack down on the stock trading manipulation.

Key wordsIntraday Trading Weighted Network    Logistic Regression    Stock Trading Manipulation    Discriminating Model
收稿日期: 2019-02-21     
中图分类号:  F832.5  
基金资助:*本文系国家社会科学基金项目“股票操纵网络的演化机理与治理模式研究”的研究成果之一(16BGL066)
通讯作者: 张建锋     E-mail: 527802384@qq.com
引用本文:   
扈文秀,马丽,张建锋. 基于股票日内交易加权网络的超短期股票交易型操纵识别研究 *[J]. 数据分析与知识发现, 2019, 3(10): 118-126.
Wenxiu Hu,Li Ma,Jianfeng Zhang. Identifying Ultra-short-term Market Manipulation with Stock Intraday Trading Weighted Network. Data Analysis and Knowledge Discovery, DOI:10.11925/infotech.2096-3467.2019.0192.
链接本文:  
http://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2019.0192
图1  股票日内交易加权网络构建图
图2  000876被操纵期间股票日内交易加权网络拓扑结构
图3  000876未被操纵期间股票日内交易加权网络拓扑结构
变量 样本数 均值 标准差 最小值 最大值
连线数量X2 374 53 426.32 33 981.46 9520 145376
加权平均节点度X3 374 6 264.68 18 360.78 59.84 109 965.70
网络密度X4 374 0.55 0.79 0.08 3.45
网络直径X5 374 196.48 158.64 48 773
加权平均距离X6 374 20.26 15.85 7.55 94.03
最大组元节点个数X8 374 2 768.48 3 024.37 285 13225
最大组元规模X9 374 0.11 0.09 0.01 0.41
岛屿个数X11 374 3 249.61 2 072.37 350 11013
一模网加权平均点度X13 374 34.27 32.98 7.50 151.60
一模网不带环密度X14 374 0.33 0.23 0.05 1.09
一模网加权平均距离X15 374 22.97 17.89 8.69 106.15
网络度数中心势X17 374 0.07 0.11 0.001 0.49
虚拟变量D 374 0.43 0.49 0 1
表2  股票日内交易加权网络参数描述性统计
V Coef. Std. Err. z P>z [95% Conf.Interval]
连线数量X2 -0.000 049 0.000 025 -1.980 0.0 470 -0.00 009 -0.000 001
加权平均节点度X3 -0.000 764 0.000 198 -3.860 0.0 000 -0.00 115 -0.000 376
网络密度X4 0.849 989 0.365 506 2.330 0.0 200 0.13 361 1.566 368
网络直径X5 -0.000 178 0.002 367 -0.080 0.9 400 -0.00 481 0.004 461
加权平均距离X6 -18.44 234 2.742 283 -6.730 0.0 000 -23.8 171 -13.06 756
最大组元节点个数X8 -0.000 156 0.000 247 -0.630 0.5 260 -0.00 064 0.000 327
最大组元规模X9 24.16 735 11.11 035 2.180 0.0 300 2.39 146 45.94 323
岛屿个数X11 0.001 018 0.000 363 2.810 0.0 050 0.00 031 0.001 728
一模网加权平均点度X13 -0.000 660 0.014 903 -0.040 0.9 650 -0.02 987 0.028 550
一模网不带环密度X14 -6.823 294 3.132 357 -2.180 0.0 290 -12.9 626 -0.683 988
一模网加权平均距离X15 16.32 039 2.429 062 6.720 0.0 000 11.5 595 21.08 126
网络度数中心势X17 14.05 585 3.342 419 4.210 0.0 000 7.50 483 20.60 687
虚拟变量D -0.691 979 0.792 088 -0.870 0.3 820 -2.24 444 0.860 486
常数 -0.935 118 0.982 955 -0.950 0.3 410 -2.86 167 0.991 439
表3  股票交易型操纵识别加权网络参数二元逻辑回归结果
样本内检验结果 样本外检验结果
实际值 检验值 准确率(%) 实际值 检验值 准确率(%)
Y Y
1 0 1 0
Y 1 170 15 91.89 Y 91 11 89.22
0 9 180 95.24 16 102 86.44
总体准确率(%) 93.58 总体准确率(%) 87.73
表4  股票交易型操纵识别模型样本内和样本外检验结果
[1] Ögüt H, Doganay M, Aktas R . Detecting Stock-price Manipulation in an Emerging Market: The Case of Turkey[J]. Expert Systems with Applications, 2009,36(9):11944-11949.
[2] Diaz D, Theodoulidis B, Sampaio P . Analysis of Stock Market Manipulations Using Knowledge Discovery Techniques Applied to Intraday Trade Prices[J]. Expert Systems with Applications, 2011,38(10):12757-12771.
[3] Comerton-Forde C, Putniņš T J . Measuring Closing Price Manipulation[J]. Journal of Financial Intermediation, 2011,20(2):135-158.
[4] Adamic L, Brunetti C, Harris J H , et al. Trading Networks[J]. Econometrics Journal, 2017,20(3):S126-S149.
[5] Maxim M R, Ashif A S M . A New Method of Measuring Stock Market Manipulation Through Structural Equation Modeling (SEM)[J]. Investment Management and Financial Innovations, 2017,14(3):54-61.
[6] 李志辉, 周谧 . 中国股票市场操纵行为测度与影响因素研究——基于上市公司特征角度[J]. 中央财经大学学报, 2018(12):25-36.
( Li Zhihui, Zhou Mi . Research on Measurement and Influencing Factors of Manipulation in China’s Stock Market: Based on the Characteristics of Listed Companies[J]. Journal of Central University of Finance & Economics, 2018(12):25-36.)
[7] 钟廷勇, 李江娜, 郭志刚 . 股价波动、市场操纵与证券市场监管[J]. 管理世界, 2017(7):171-172.
( Zhong Tingyong, Li Jiangna, Guo Zhigang . Stock Price Fluctuation, Market Manipulation and Securities Market Regulation[J]. Management World, 2017(7):171-172.)
[8] 张建锋, 付强, 杜金柱 . 基于Logistic模型的上市公司股票价格操纵预判研究[J]. 西安理工大学学报, 2018,34(2):240-245.
( Zhang Jianfeng, Fu Qiang, Du Jinzhu . Study on the Manipulation Anticipation of Listed Companies’ Stock Prices Based on Logistic Model[J]. Journal of Xi’an University of Technology, 2018,34(2):240-245.)
[9] 李志辉, 王近, 李梦雨 . 中国股票市场操纵对市场流动性的影响研究——基于收盘价操纵行为的识别与监测[J]. 金融研究, 2018(2):135-152.
( Li Zhihui, Wang Jin, Li Mengyu . A Study on China’s Stock Market Manipulation’s Effects on Market Liquidity: Based on Closing Price Manipulation Behavior’s Identification and Monitoring[J]. Journal of Financial Research, 2018(2):135-152.)
[10] Sun X, Cheng X, Shen H , et al. Distinguishing Manipulated Stocks via Trading Network Analysis[J]. Physica A: Statistical Mechanics and Its Applications, 2011,390(20):3427-3434.
[11] Jiang Z, Xie W, Xiong X , et al. Trading Networks Abnormal Motifs and Stock Manipulation[J]. Quantitative Finance Letters, 2013,1(1):1-8.
[12] Sun X, Shen H, Cheng X , et al. Detecting Anomalous Traders Using Multi-slice Network Analysis[J]. Physica A: Statistical Mechanics and Its Applications, 2017,473:1-9.
[13] Shi F, Sun X, Shen H , et al. Detect Colluded Stock Manipulation via Clique in Trading Network[J]. Physica A: Statistical Mechanics and Its Applications, 2019,513:565-571.
[14] Ding C, Yao H, Du J , et al. Cascading Failure in Interconnected Weighted Networks Based on the State of Link[J]. International Journal of Modern Physics C, 2017,28(3):1750040.
[1] 陈先来,韩超鹏,安莹,刘莉,李忠民,杨荣. 基于互信息和逻辑回归的新词发现 *[J]. 数据分析与知识发现, 2019, 3(8): 105-113.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn