连续双向拍卖市场中交易策略对价格行为的影响研究
|
摘要
连续竞价（又称连续双向拍卖）和集合竞价是世界大多数国家金融证券市场中普遍采用的交易机制。长期以来，人们对股票市场价格行为的研究，一直将股票交易过程视为一个“黑盒”，即当相关信息输入后，就会自动产生一个有效的定价，市场则根据这一价格为导向实现资本的最优配置。但是，近年来随着实验经济学和金融市场微观结构理论的兴起，人们正试图打开这一“黑盒”，研究交易者的交易行为对市场价格行为的影响。交易策略决定交易者行为，交易者选取的交易策略不同，对价格行为的影响肯定不同，本文通过选取几种典型的交易策略来研究这一问题。
本文运用计算机仿真的方法，在模拟的连续双向拍卖市场中研究交易策略对价格行为的影响。我们选取了连续双向拍卖市场中两种具有代表性的交易策略为研究对象，即有约束“零信息”交易策略（ZI-C）和增强“零信息”交易策略（ZIP）。首先，分别研究了纯ZI-C策略和纯ZIP策略下，价格行为的统计特征和时间序列特征差异；然后，选取不同比例的ZI-C交易策略和ZIP交易策略进行仿真实验，研究发现：当采取某种策略的比例较大时，价格行为特征则更接近所有交易者均采取该策略时的价格行为特征。因此，得出交易行为决定了价格行为的结论。由于连续双向拍卖市场是现代金融交易所的最重要的交易机制，本文所做的工作对研究金融市场的价格行为有一定的意义。
Continuous Double Auction and Call Market are the trading mecha nisms that thefina ncia l markets of most nations adapt. Ever since a long time ago，we study the tradingprocess of stock trading as a black box which yields an effectual price if some informationis input, when we research the price behavior of stock market, and after that the marketmake the capital alloca tion optima l. As the experimental economics and MicrostructureTheory develop, the scholars are trying to open the black box by the research about theaffection of trading behavior to price behavior in the double auction market. Price strategyis an important part of trading behavior，different strategy has different affection. Thispassage selects severa l typica l trading strategies to study this problem.
In a computerized continuous double auction CDA market with Zero-intelligence withconstraint ZI-C agents and Zero-intelligence plus ZIP agents, this paper studies the pricebehavior of the CDA market with homogenous ZI-C agents, homogenous ZIP agents, andheterogeneous agents of the two. Regular patterns of price behaviors with the twohomogenous agents are provided by time series analysis and statistic analysis. Thetransformation pattern of price behavior is analyzed with different percent ofheterogeneous agents of the two. It is that when there are more agents applying the samestrategy, the price behavior gets closer to the pattern with that homogenous strategy. Forthe reason that the CDA is the premier market format in most modern fina ncia l excha nges,the find ing is useful to study the price behavior in the fina nce market.
本文运用计算机仿真的方法，在模拟的连续双向拍卖市场中研究交易策略对价格行为的影响。我们选取了连续双向拍卖市场中两种具有代表性的交易策略为研究对象，即有约束“零信息”交易策略（ZI-C）和增强“零信息”交易策略（ZIP）。首先，分别研究了纯ZI-C策略和纯ZIP策略下，价格行为的统计特征和时间序列特征差异；然后，选取不同比例的ZI-C交易策略和ZIP交易策略进行仿真实验，研究发现：当采取某种策略的比例较大时，价格行为特征则更接近所有交易者均采取该策略时的价格行为特征。因此，得出交易行为决定了价格行为的结论。由于连续双向拍卖市场是现代金融交易所的最重要的交易机制，本文所做的工作对研究金融市场的价格行为有一定的意义。
Continuous Double Auction and Call Market are the trading mecha nisms that thefina ncia l markets of most nations adapt. Ever since a long time ago，we study the tradingprocess of stock trading as a black box which yields an effectual price if some informationis input, when we research the price behavior of stock market, and after that the marketmake the capital alloca tion optima l. As the experimental economics and MicrostructureTheory develop, the scholars are trying to open the black box by the research about theaffection of trading behavior to price behavior in the double auction market. Price strategyis an important part of trading behavior，different strategy has different affection. Thispassage selects severa l typica l trading strategies to study this problem.
In a computerized continuous double auction CDA market with Zero-intelligence withconstraint ZI-C agents and Zero-intelligence plus ZIP agents, this paper studies the pricebehavior of the CDA market with homogenous ZI-C agents, homogenous ZIP agents, andheterogeneous agents of the two. Regular patterns of price behaviors with the twohomogenous agents are provided by time series analysis and statistic analysis. Thetransformation pattern of price behavior is analyzed with different percent ofheterogeneous agents of the two. It is that when there are more agents applying the samestrategy, the price behavior gets closer to the pattern with that homogenous strategy. Forthe reason that the CDA is the premier market format in most modern fina ncia l excha nges,the find ing is useful to study the price behavior in the fina nce market.
引文
[1]詹文杰，双向拍卖的交易机制和交易策略[M].武汉：华中科技大学出版社.2008,1-3.
[2]陈保华.交易机制对股价行为的影响-对中国股票市场的实证检验[J].经济研究,2001(5): 69-73, 95.
[3] Fried man D. A Simple Testable Model of Double Auction Markets [J]. Journal ofEconomic Beha vior and Organiza tion, 1991, 15(1): 47-70
[4] Gode D K, Sunder S. Double Auction Dynamics: Structural Consequences of Non-Binding Price Controls [J]. Journal of Economic Dynamics and Control, 2004, 28(7):1707-1731.
[5]詹文杰,汪寿阳.从“Smith’s奥秘”评双向拍卖的研究进展[J].管理科学学报,2003, 6(1): 1-12.
[6] Cardoso H.L., Schaefer M., Oliveira E. A Multi-Agent System for ElectronicCommerce Includ ing Adaptive Strategic Beha viours [C]. Lecture Notes in ComputerScience (EPIA’99), Berlin: Springer, 1999: 252-266.
[7] Tesauro G, Das R. High-performa nce Bidd ing Agents for the Continuous DoubleAuction [C]. Proceedings of the 3rd ACM conference on Electronic Commerce, Tampa:2001,48-54.
[8] Brandouy O, Barneto P, Leger L. Insider Trading. Imitative Beha vior and PriceFormation in a Simula ted Double-Auction Stock Market [R]. Economic ResearchPaper No.00/01, Loughborough University, UK, 2000,99-104.
[9] Duffy J. Learning to speculate: Experiments with artificia l and real agents [J]. Journalof Economic Dynamics and Control, 2001, 25: 295-319.
[10] Plott C R, Smith V L. An Experimental Examination of Two Excha nge Institutions [J].Review of Economic Studies, 1978, 45 (1): 133-153.
[11]刘海龙，郑立辉，吴冲锋.现代金融理论的进展综述[J].系统工程理论与实践，2001，21(1):14-20.
[12]付静,邵培基,杨小平.在线双向拍卖中的不完全信息博弈仿真研究[J].管理学报, 2006, 3(6): 673-676, 682.
[13] Madhava n.A.Consolidation,fragmentation,a nd the disclo-sure of tradinginformation[J].Reviewof Financia l Studies, 1995, 579-603.
[14]何杰.证券市场微观结构理论[J].经济导刊，2000，5:31-38.
[15] Cason T N, Fried man D. Price Formation in Double Auction Markets [J]. Journal ofEconomic Dynamics and Control, 1996, 20(8): 1307-1337.
[16] Dawid H. Adaptive Learning by Genetic Algorithms: Analytical Results andApplication to Economic Models [M] //Lecture Notes in Economics and Mathema ticalSystems, Berlin: Springer, 1996, 69-75.
[17] He M, Leung H, Jennings, N. A Fuzzy Logic Based Bidd ing Strategy for AutonomousAgents in Continuous Double Auctions [J]. IEEE Transactions on Knowled ge andData Engineering, 2003, 15(6): 1345-1363.
[18]张维,刘文财,王启文等.面向资本市场复杂性建模: Agent计算实验金融学[J] .现代财经, 2003, 23(1): 3-7.
[19] Fama E, French K. Multifactor Expla nation of Asset Pricing Anoma lies [J]. Journalof Finance, 1996, 51(1): 55-84.
[20]刘海龙,吴冲锋.金融市场微观结构理论综述[J].管理评论, 2003, 15(1): 38-40.
[21]萨顾岚.中国股票市场实证统计分析[M].北京:中国财政经济出版社, 1999: 40-50.
[22]吴林祥.证券交易制度分析[M].上海:上海财经大学出版社, 2002.114-119.
[23]穆启国,刘海龙,吴冲锋.涨跌幅限制与股票价格行为分析[J].管理科学学报,2004(6): 23-30.
[24] LeBaron B. Agent-based computationa l fina nce: Suggested readings and earlyresearch [J]. Journal of Economic Dynamics and Control, 2000, 24(5-7): 679-702.
[25]俞乔，市场有效、周期异动与股价波动--对上海、深圳股票市场的实证分析[J].经济研究,994(9): 43-50.
[26]张思奇,马刚,冉华.股票市场风险、收益与市场效率: ARMA-ARCH-M模型[J].世界经济, 2000(5): 19-28.
[27]李学,欧阳俊,秦宛顺.中国股市的星期效应研究[J].统计研究, 2001(8): 38-41.
[28]陈超,钱苹.中国股票市场“周内效应”再检验[J].经济科学, 2002(1): 85-91.
[29]刘建军,李辉.我国股市有效性的统计检验[J].江苏统计, 2002(11): 21-23.
[30] Barberis N, Shleifer A, Vishny R. A Model of Investor Sentiment [J]. Journal ofFinancia l Economics, 1998, 49(9): 307–343.
[31] Daniel K, Hirshleifer D, Subrahma nyam A. Investors, Psychology and SecurityMarket Under-and overreactions [J]. Journal of Finance, 1998, 53(6): 1839–1885.
[32]林文.行为金融理论文献综述[J].云南财贸学院学报:社会科学版, 2004, 3: 32-34.
[33] Barberis N, Shleifer A, Vishny R. A Model of Investor Sentiment [J]. Journal ofFinancia l Economics, 1998, 49(9): 307–343.
[34] Daniel K, Hirshleifer D, Subrahma nyam A. Investors, Psychology and SecurityMarket nder-and overreactions [J]. Journal of Finance, 1998, 53(6): 1839–1885.
[35] Hong H, Stein J. A Unified Theory of Under-reaction, Momentum Trading andOverreaction in Asset Markets [J]. Journal of Finance, 1999, 54(6): 2143–2184.
[36] Fried man L. A competitive bidding strategy [J]. Operation Research, 1956, (4): 104-112.
[36]詹文杰.基于实验经济学的双向拍卖研究[D].武汉:华中科技大学博士学位论文, 2002,11-15.
[37] Zhan W, Wang S, Zha ng J, Yang J, Lai K K. Marsha llia n Deviation: New ObservableCriterion to Measure Transaction Path for k-ZI traders in Double Auction Markets [J].Journal of Systems Science and Complexity, 2002, 15(3): 261-277.
[38] Zhan W, Wang S, Zha ng J, Yang J, Lai K K. Rela tionship between Offer Strategy andTrade Ratio for k-ZI Traders in Continuous Double Auction Market [J]. Internationa lJournal of Information Technology and Decision Making, 2003, 2(3): 381-395.
[39] Zhan W, Yang J. Strategy and Endowment Effects of k-ZI Strategy in ContinuousDouble Auction Markets [C]. The 4th IEEE Internationa l conference on WirelessCommunica tions, Networking and Mobile Computing (WiCom2008), Dalia n, 2008，224-231.
[40]詹文杰,张金隆,蔡铂.双向拍卖的报价决策模型研究[J].统计与决策, 2002,128(6): 29-30.
[41]詹文杰,张金隆,杨洁.双向拍卖中的“零信息”模型研究[J].科技进步与对策,2002, 19(6): 108-109.
[42] Zhan W, Wang S, Zha ng J, Yang J, Lai K K. k-ZI: A General Zero-Intelligence Modelfor Continuous Double Auction Markets [J]. Internationa l Journal of InformationTechnology and Decision Making, 2002, 1(4): 673-691.
[43] Zhan W, Wang Y. Effect of k-ZI trader’s trading strategy on performa nce of continuousdouble auction market [C]. Proceedings of 13th Internationa l Conference onManagement Science and Engineering, Lille, France, 2006: 271-275.
[44]詹文杰,占俊英.连续双向拍卖中的“增强零信息”策略研究[J].管理评论, 2008,20(5): 44-50.
[45] Zhan W, Wang S, Lai K K. Competitive Analysis of Augmented Zero-IntelligenceAgents for Double Auction Markets [C]. Proceedings of the First Internationa lWorkshop on Competitive Bidd ing and Auctions, Beijing, 2002: 246-269.
[46] Gjerstad S, Dick haut J. Price Formation in Double Auctions [J]. Games and EconomicBeha vior, 1992, 22:1-29.
[47] Sunju P., Durfee E.H., Willia m P.B. An Adaptive Agent Bidd ing Strategy Based onStochastic Modeling. Proceedings of the Third Internationa l Conference onAutonomous Agents (Agents'99), 1999，111-117.
[48]盛骤.概率论与数理统计[M].高等教育出版社，1997，215-220.
[49]高铁梅.计量分析与建模[M].清华大学出版社，2005，85-91.
[50] Zhan W, Song H. How Trading Strategy determine Price Beha vior in ContinuousDouble Auction Market [C]. Proceeding of Internationa l Conference on businessintelligence and fina ncial engineering (BIFE’2008), Changsha, 2008.1045-1052
[2]陈保华.交易机制对股价行为的影响-对中国股票市场的实证检验[J].经济研究,2001(5): 69-73, 95.
[3] Fried man D. A Simple Testable Model of Double Auction Markets [J]. Journal ofEconomic Beha vior and Organiza tion, 1991, 15(1): 47-70
[4] Gode D K, Sunder S. Double Auction Dynamics: Structural Consequences of Non-Binding Price Controls [J]. Journal of Economic Dynamics and Control, 2004, 28(7):1707-1731.
[5]詹文杰,汪寿阳.从“Smith’s奥秘”评双向拍卖的研究进展[J].管理科学学报,2003, 6(1): 1-12.
[6] Cardoso H.L., Schaefer M., Oliveira E. A Multi-Agent System for ElectronicCommerce Includ ing Adaptive Strategic Beha viours [C]. Lecture Notes in ComputerScience (EPIA’99), Berlin: Springer, 1999: 252-266.
[7] Tesauro G, Das R. High-performa nce Bidd ing Agents for the Continuous DoubleAuction [C]. Proceedings of the 3rd ACM conference on Electronic Commerce, Tampa:2001,48-54.
[8] Brandouy O, Barneto P, Leger L. Insider Trading. Imitative Beha vior and PriceFormation in a Simula ted Double-Auction Stock Market [R]. Economic ResearchPaper No.00/01, Loughborough University, UK, 2000,99-104.
[9] Duffy J. Learning to speculate: Experiments with artificia l and real agents [J]. Journalof Economic Dynamics and Control, 2001, 25: 295-319.
[10] Plott C R, Smith V L. An Experimental Examination of Two Excha nge Institutions [J].Review of Economic Studies, 1978, 45 (1): 133-153.
[11]刘海龙，郑立辉，吴冲锋.现代金融理论的进展综述[J].系统工程理论与实践，2001，21(1):14-20.
[12]付静,邵培基,杨小平.在线双向拍卖中的不完全信息博弈仿真研究[J].管理学报, 2006, 3(6): 673-676, 682.
[13] Madhava n.A.Consolidation,fragmentation,a nd the disclo-sure of tradinginformation[J].Reviewof Financia l Studies, 1995, 579-603.
[14]何杰.证券市场微观结构理论[J].经济导刊，2000，5:31-38.
[15] Cason T N, Fried man D. Price Formation in Double Auction Markets [J]. Journal ofEconomic Dynamics and Control, 1996, 20(8): 1307-1337.
[16] Dawid H. Adaptive Learning by Genetic Algorithms: Analytical Results andApplication to Economic Models [M] //Lecture Notes in Economics and Mathema ticalSystems, Berlin: Springer, 1996, 69-75.
[17] He M, Leung H, Jennings, N. A Fuzzy Logic Based Bidd ing Strategy for AutonomousAgents in Continuous Double Auctions [J]. IEEE Transactions on Knowled ge andData Engineering, 2003, 15(6): 1345-1363.
[18]张维,刘文财,王启文等.面向资本市场复杂性建模: Agent计算实验金融学[J] .现代财经, 2003, 23(1): 3-7.
[19] Fama E, French K. Multifactor Expla nation of Asset Pricing Anoma lies [J]. Journalof Finance, 1996, 51(1): 55-84.
[20]刘海龙,吴冲锋.金融市场微观结构理论综述[J].管理评论, 2003, 15(1): 38-40.
[21]萨顾岚.中国股票市场实证统计分析[M].北京:中国财政经济出版社, 1999: 40-50.
[22]吴林祥.证券交易制度分析[M].上海:上海财经大学出版社, 2002.114-119.
[23]穆启国,刘海龙,吴冲锋.涨跌幅限制与股票价格行为分析[J].管理科学学报,2004(6): 23-30.
[24] LeBaron B. Agent-based computationa l fina nce: Suggested readings and earlyresearch [J]. Journal of Economic Dynamics and Control, 2000, 24(5-7): 679-702.
[25]俞乔，市场有效、周期异动与股价波动--对上海、深圳股票市场的实证分析[J].经济研究,994(9): 43-50.
[26]张思奇,马刚,冉华.股票市场风险、收益与市场效率: ARMA-ARCH-M模型[J].世界经济, 2000(5): 19-28.
[27]李学,欧阳俊,秦宛顺.中国股市的星期效应研究[J].统计研究, 2001(8): 38-41.
[28]陈超,钱苹.中国股票市场“周内效应”再检验[J].经济科学, 2002(1): 85-91.
[29]刘建军,李辉.我国股市有效性的统计检验[J].江苏统计, 2002(11): 21-23.
[30] Barberis N, Shleifer A, Vishny R. A Model of Investor Sentiment [J]. Journal ofFinancia l Economics, 1998, 49(9): 307–343.
[31] Daniel K, Hirshleifer D, Subrahma nyam A. Investors, Psychology and SecurityMarket Under-and overreactions [J]. Journal of Finance, 1998, 53(6): 1839–1885.
[32]林文.行为金融理论文献综述[J].云南财贸学院学报:社会科学版, 2004, 3: 32-34.
[33] Barberis N, Shleifer A, Vishny R. A Model of Investor Sentiment [J]. Journal ofFinancia l Economics, 1998, 49(9): 307–343.
[34] Daniel K, Hirshleifer D, Subrahma nyam A. Investors, Psychology and SecurityMarket nder-and overreactions [J]. Journal of Finance, 1998, 53(6): 1839–1885.
[35] Hong H, Stein J. A Unified Theory of Under-reaction, Momentum Trading andOverreaction in Asset Markets [J]. Journal of Finance, 1999, 54(6): 2143–2184.
[36] Fried man L. A competitive bidding strategy [J]. Operation Research, 1956, (4): 104-112.
[36]詹文杰.基于实验经济学的双向拍卖研究[D].武汉:华中科技大学博士学位论文, 2002,11-15.
[37] Zhan W, Wang S, Zha ng J, Yang J, Lai K K. Marsha llia n Deviation: New ObservableCriterion to Measure Transaction Path for k-ZI traders in Double Auction Markets [J].Journal of Systems Science and Complexity, 2002, 15(3): 261-277.
[38] Zhan W, Wang S, Zha ng J, Yang J, Lai K K. Rela tionship between Offer Strategy andTrade Ratio for k-ZI Traders in Continuous Double Auction Market [J]. Internationa lJournal of Information Technology and Decision Making, 2003, 2(3): 381-395.
[39] Zhan W, Yang J. Strategy and Endowment Effects of k-ZI Strategy in ContinuousDouble Auction Markets [C]. The 4th IEEE Internationa l conference on WirelessCommunica tions, Networking and Mobile Computing (WiCom2008), Dalia n, 2008，224-231.
[40]詹文杰,张金隆,蔡铂.双向拍卖的报价决策模型研究[J].统计与决策, 2002,128(6): 29-30.
[41]詹文杰,张金隆,杨洁.双向拍卖中的“零信息”模型研究[J].科技进步与对策,2002, 19(6): 108-109.
[42] Zhan W, Wang S, Zha ng J, Yang J, Lai K K. k-ZI: A General Zero-Intelligence Modelfor Continuous Double Auction Markets [J]. Internationa l Journal of InformationTechnology and Decision Making, 2002, 1(4): 673-691.
[43] Zhan W, Wang Y. Effect of k-ZI trader’s trading strategy on performa nce of continuousdouble auction market [C]. Proceedings of 13th Internationa l Conference onManagement Science and Engineering, Lille, France, 2006: 271-275.
[44]詹文杰,占俊英.连续双向拍卖中的“增强零信息”策略研究[J].管理评论, 2008,20(5): 44-50.
[45] Zhan W, Wang S, Lai K K. Competitive Analysis of Augmented Zero-IntelligenceAgents for Double Auction Markets [C]. Proceedings of the First Internationa lWorkshop on Competitive Bidd ing and Auctions, Beijing, 2002: 246-269.
[46] Gjerstad S, Dick haut J. Price Formation in Double Auctions [J]. Games and EconomicBeha vior, 1992, 22:1-29.
[47] Sunju P., Durfee E.H., Willia m P.B. An Adaptive Agent Bidd ing Strategy Based onStochastic Modeling. Proceedings of the Third Internationa l Conference onAutonomous Agents (Agents'99), 1999，111-117.
[48]盛骤.概率论与数理统计[M].高等教育出版社，1997，215-220.
[49]高铁梅.计量分析与建模[M].清华大学出版社，2005，85-91.
[50] Zhan W, Song H. How Trading Strategy determine Price Beha vior in ContinuousDouble Auction Market [C]. Proceeding of Internationa l Conference on businessintelligence and fina ncial engineering (BIFE’2008), Changsha, 2008.1045-1052