摘要
传统的金融理论往往假定投资者是完全理性同质的,但是投资者异质性和有限理性是不争的事实。随着金融市场异象的不断发现,关于投资者完全理性的假设受到越来越多的挑战。投资者的有限理性和动态学习变化适应性往往会对股票市场产生很多影响。
金融系统是一种复杂性适应性系统。股票市场中有大量不同投资者,各个投资者的思维习惯、主观偏好以及行为方式等截然不同,并且交易者的这些特征又是不断发展变化的。市场中的个体在这种持续不断的交互过程中不断地学习并根据学习得到经验改变自身结构和行为方式；各个底层个体通过相互作用,在整体层次上涌现出新的结构和更复杂的行为,本文正是基于这种思想考察了在金融市场上社会学习行为的影响。
基于agent的计算实验金融学是随市场微观结构、行为金融学、复杂性系统理论和计算机技术不断发展而交叉形成的一种自底而上的研究方法,通过复杂性系统涌现等方法研究金融市场的问题。本文以圣塔菲研究所的人工股票市场为基础构建了基于交易者社会学习机制的人工股票市场。设定多个异质交易者,交易者通过社会学习实现进化和适应环境。每个交易者的交易规则共同组成公共规则集,公共规则集的演化和学习靠遗传算法来实现。本文分析和研究了在不同学习速度下,交易者社会交互学习行为对金融市场宏观层面和交易者微观层面的影响。得到了如下结论：社会学习是引起股票价格大幅波动的原因之一,社会学习不能使股票价格收敛于有效市场假说下的理论价格,模拟出了与真实股票市场相类似的一些典型特征如股票价格的尖峰肥尾性、波动聚集性等。并在微观层面上分析了交易者的股票持有量和财富水平等。
It is often assumed that investors are completely rational and homogeneous under traditional financial theory. But more and more anomalies in security market doubt the assumption that investors are perfect rationality. The investor's heterogeneity and bounded rationality is an indisputable fact. Bounded rationality and dynamic learning of investors the stock market change adaptation will often produce a lot of impact.
Financial system is a complex adaptive system. There are a number of different heterogeneous investors in the stock market. The habit of thinking, subjective preferences and behavior way of investors are different. And these features of investors are constantly changing. In this ongoing process of continuous interactive learning and learning by experience, the individual in stock market changes his own structure and behavior. Underlying interaction of each individual emerged new structures and more complex behavior in the overall level. This article examines the impact of social learning in the financial market based on this idea of complex adaptive system.
With the development of computer technology and finance theory agent-based computational finance becomes a new research field of finance. It is based on theories including the finance market microstructure, behavioral finance and complex systems theory. Study on financial issues makes use of emergence methods of complex systems. In this paper, a new artificial stock market (ASM) in which investors have the ability of social learning is built, based on Santa Fe Institute artificial stock market. In this artificial stock market, the heterogeneous investors evolve and adapt to the environment through social learning. Public rule set is composed of the transaction rule of each investor. And the public rule set evolves by genetic algorithm. This paper analyzes and researches the impact of social learning in the different learning speeds on financial market and micro-level investor. In this paper the following conclusions are obtained. Social learning is one of the reasons of the stock price volatility. Stock price in the artificial stock market can't converge to the theoretical price under the efficient market hypothesis. And the artificial stock market simulates some typical characteristics of the real stock market. These typical characteristics include the spike and fat tail of stock price, and volatility clustering of stock price. And stock holdings and wealth levels of investors are analyzed in the micro-level.
引文
[1]Andrei Shleifer, Inefficient Markets:An Introduction to Behavioral Finance, Oxford University Press,2000
[2]W.B.Arthur,J.H.Holland,B.LeBaron,et al.Asset Pricing under Endogenous Expectations in an Artificial Stock Market.In W.B.Arthur,S. Durlauf and D.Lane,eds, The Economy as an Evolving Complex System 2, Addison-Wesley, 1997:15-44
[3]B LeBaron, W B Arthur, R Palmer. Time series properties of an artificial stock market. Journal of Economic Dynamics & Control,1999,23:487-516
[4]T.Lux.Herd Behavior, Bubbles and Crashes.Economics Journal.1995, 105(431):881-896
[5]T.Lux.The Socio-economic Dynamics of Speculative Markets:Interacting Agents, Chaos, and the Fat Tails of Return Distributions.Journal of Economic Behavior and Organization.1998,33(2):143-165
[6]Xavier Vives. Social learning and rational expectations. European Economic Review,1996,40(4):589-601
[7]T.Lux, M.Marchesi.Scalling and Criticality in a Stochastic Multi-agent Model of a Financial Market.Nature.1999,397(11):498-500
[8]R.Palmer, W.B.Arthur, J.H.Holland, et al.Artificial Economic Life:A Simple Model of a Stock Market.Physica D.1994,75(1-3):264-274
[9]Banerjee, A., A Simple Model of Herd Behavior. Quarterly Journal of Economics, 1992,107:797-817
[10]Bikhchandani, S., Hirshleifer, D.&Welch, I., A Theory of Fads, Fashion, Custom and Cultural Change as Informational Cascades.Journal of Political Economy,1992,100:992-1026
[11]Campbell J Y, Shiller R J.Cointegration and tests of present value model.The Journal of Political Economy,1987,95:1062-1088
[12]Chen S H, Yeh C. Evolving Traders and the Business School with Genetic Programming:A New Architecture of Agent-based Artificial Stock Market. Journal of Economic Dynamics and Control,2001, (25):363-394
[13]Kendall Q Su Y., The Co-evolution of Trading Strategies in A Multi-agent Based Simulated Stock Market Through the Integration of Individual Learning and Social Learning. Evolutionary Computation,2003, (4):2298-2305
[14]Grossman S, Stiglitz F.On the Impossibility of Informationally Efficient Markets. American Economic Review,1980,70:393-408
[15]Chen S H, Yeh C H.On the emergent properties of artificial stock markets. Journal of Economic Behavior and Organization,2002,49(2):217-239
[16]LeBaron B.Evolution and time horizons in an agent-based stock market.Macroeconomic Dynamics,2001,5(2):225-254
[17]Kahneman D., and A.Tversky.Prospect Theory:An Analysis of Decision Making under Risk.Econometrica,1979,47(2):263-291
[18]De Bondt W., and R.Thaler.Does the Stock Market Overreact?.Journal of Finance,1985,40:793-808
[19]Bray M. Learning, estimation, and the stability of expectations. Journal of Economic Theory,1982,26:318-339
[20]Shefrin H., and M.Statman. Behavioral Capital Asset Pricing Theory.Journal of Finance and Quantitative Analysis,1994,29:323-349
[21]Shefrin H., and M.Statman.Behavioral Portfolio Theory.Journal of Finance and Quantitative Analysis,2000,35:127-151
[22]BarBeris N., A.Shleifer R. Vishny.A Model of Investor Sentiment, Journal of Financial Economics.1998,49:307-343
[23]Daniel, K., Hirshleifer D., Subrahmanyam, A.Investor Psychology and Security Market Under-and Overreactions. Journal of Finance.1998.53:1839-1885
[24]Hong H., J.C.Stein. A Unified Theory of Underreaction, Momentum Trading, and Overreaction in Asset Markets. Journal of Finance.1999.54:2143-2184
[25]N.J.Vriend.An illustration of the essential difference between individual and social learning, and its consequences for computational analyses.Journal of Economic Dynamics and Control,,2000,24(1):23-30
[26]Palmer R, Arthur W B, Holland J H,et al. Artificial economic life:A simple model of a stock market.Physica D,1994,75:264-274.
[27]Duffy J. Agent-based models and human-subject experiments [A]. Tesfatsion, Judd K L. Handbook of Computational Economics, Volume 2 Agent-Based Computational Economics, Handbooks in Economics Series[C]. North-Holland: Elsevier,2006.
[28]Levy M, Levy H, Solomon S. Microscopic Simulation of Financial Markets. New York:Academic Press,2000.46-58.
[29]Bak P, Paczuski M, Shubik M. Price variation in a stock market with many agents. Physica A,1997,246(3-4):430-453.
[30]LeBaron B. Agent-based computational finance [C]//Judd K. Tesfatsion L Handbook of Computational Economics:Vol.2. Elsevier/North-Holland,2006: 1187-1233
[31]Lettau M. Explaining the facts with adaptive agents:the case of mutual fund flows. Journal of Economic Dynamics and Control,1997,21:1117-1148
[32]O'Hara, Maureen,1995, Market Microstructure Theory, Blackwell Publishers Inc., Cambridge, MA
[33]Brenner, T.(2006), Agent learning representation:advice on modeling economic learning, In:Handbook of Computational Economics, Vol 2:Agent-Based Computational Economics Chapter18, Edited by L.Tesfatsion and K.L.Judd Elsevier Science B.V,2006.
[34]Shu-Heng Chen, Chung-Chih Liao, Agent-based computational modeling of the stock price-volume relation, Information Sciences,2005,170(2):75-100
[35]宋逢明,林森,李超,基于人工股票市场分析持有期对投资者收益的影响,运筹与管理,2008,17(1),88-93
[36]马进胜,杨敏,邱菀华,基于主体的非均衡人工股市计算模型,北京化工大学学报,2008,35(2)：99-104
[37]刘兴华,汤兵勇,智能体建模和资本市场复杂性,管理科学学报,2005,8(4)：35-42
[38]刘成,颜钢锋,股票投资策略选择的计算机仿真,江南大学学报(自然科学版),2007,6(3)：302-305
[39]邹琳,马超群,李红权,中国股市仿真系统建模及其非线性特征研究,系统管理学报,2008,17(4)：385-389
[40]雷英杰,张善文,李续武,周创明,MATLAB遗传算法工具箱及应用,西安：西安电子科技大学出版社,2005,11-16
[41]袁毅贤,梁莹,基于Agent计算金融的计算机仿真研究综述,计算机仿真,2007,24(2)：262-265
[42]鲜于波,梅琳,经济复杂性系统主体学习算法理论综述,复杂系统与复杂性科学,2007,4(2)：77-92
[43]汪曙霞,郭晔,曾志钊,陈雨,市场微观结构：理论、实践与监管应用,北京：中国财政经济出版社,2006,15-20
[44]董志勇,行为金融学,北京：北京大学出版社,2009,7-17
[45]高宝俊,戴辉,宣慧玉,基于Agent的股票市场仿真：个体行为对市场、政策效果的影响,系统工程理论方法应用,2005,14(6)：497-501
[46]李锬,基于异质交易者的期货市场价格动态研究,博士学位论文,哈尔滨工业大学,2007
[47]杨春,有限理性、噪音交易与过度波动,博士学位论文,天津大学,2006
[48]刘大海,王治宝,孙洪军,王秀峰,基于多agent的虚拟股市仿真研究,计算机工程与应用,2003,25：26-28
[49]刘文财,刘豹,王启文,张维,基于Agent的金融市场模型研究进展综述,系统工程学报,2003,18(2)：135-141
[50]约翰·霍兰,隐秩序—适应性造就复杂性,上海：上海科技教育出版社,2000
[51]张维,赵帅特,基于ACF的行为金融研究局限及未来研究方向,2008,28(10)：9-13
[52]于同奎,多种市场形态下人工股市特征事实分析,复杂系统与复杂性科学2008,5(2)：62-71
[53]刘维妮,韩立岩,管理学报,2007,4(4)：414-420