SpringerLink
Log in

Can Artificial Traders Learn and Err Like Human Traders? A New Direction for Computational Intelligence in Behavioral Finance

  • Chapter
  • First Online:
Financial Decision Making Using Computational Intelligence

Part of the book series: Springer Optimization and Its Applications ((SOIA,volume 70))

  • 2048 Accesses

Abstract

The microstructure of markets involves not only human traders’ learning and erring processes but also their heterogeneity. Much of this part has not been taken into account in the agent-based artificial markets, despite the fact that various computational intelligence tools have been applied to artificial-agent modeling. One possible reason for this little progress is due to the lack of good-quality data by which the learning and erring patterns of human traders can be easily archived and analyzed. In this chapter, we take a pioneering step in this direction by, first, conducting double auction market experiments and obtaining a dataset involving about 165 human traders. The controlled laboratory setting then enables us to anchor the observing trading behavior of human traders to a benchmark (a global optimum) and to develop a learning index by which the learning and erring patterns can be better studied, in particular, in light of traders’ personal attributes, such as their cognitive capacity and personality. The behavior of artificial traders driven by genetic programming (GP) is also studied in parallel to human traders; however, how to represent the observed heterogeneity using GP remains a challenging issue.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
EUR 29.95
Price includes VAT (Germany)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
EUR 85.59
Price includes VAT (Germany)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
EUR 106.99
Price includes VAT (Germany)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free ship** worldwide - see info
Hardcover Book
EUR 106.99
Price includes VAT (Germany)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free ship** worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    Various patterns of mistakes, also known as behavioral biases, have been long studied by psychologists and social scientists. See [5], Part II, for a review of various biases. Also see [34].

  2. 2.

    Learning does not necessarily mean correction in a right direction; the well-known over-reaction or over-adjustment are typical examples of this pattern of learning [48]. Furthermore, learning may take a while to see its effect; this is known as slower learning or the inertial effect [12].

  3. 3.

    See [33] for a simple historical review of the use of this term.

  4. 4.

    For example, the recently published handbook on metaheuristics [31] has no single mention of psychology.

  5. 5.

    Probably partially because of this gap, artificial traders cannot replace human traders [15].

  6. 6.

    Recently, there have been a number of studies focusing on the neurocognitive study of decision making under uncertainty or ambiguity, which may well serve as a neural foundation for the observed behavioral phenomena here [36, 54].

  7. 7.

    Chen [17] argues that genetic programming equips economists with a tool to model the chance-discovering agent, which is an essential element of modern economic theory.

  8. 8.

    The details of the GP run in this chapter can be found in [22].

  9. 9.

    The fundamental pursuit here is: when a mistake is made, what are the differences between that made by an artificial agent and that made by human agents?

  10. 10.

    It does not have to be narrowly limited to the pecuniary costs associated with trading, such as broker fees or the Tobin tax. It can cost personal health as well [6].

  11. 11.

    In fact, there are a total of 185 subjects attending the double auction experiments, but for some of them the data are incomplete. Hence, for the WMC test the valid sample has 165 subjects, and for the personality test the valid sample has 168 subjects. There are 151 subjects appearing in both samples.

  12. 12.

    For those readers who are unfamiliar with this development, some backgrounds are available from [19, 20].

  13. 13.

    While the conversation between psychology and economics has a long history and a rapidly growing literature, it was only very recently that economists started to take into account psychological attributes in their economic modeling and analysis.

  14. 14.

    A survey is available from [21].

  15. 15.

    From Fig. 2.7, for a trader who is identified as a case of learning the optimum in period 3, his learning index must be in plateau A in the first three periods. Actually, Subject 1531 started performing the optimal strategy in period 2 until the end of the experiment except for one period obviously due to a typo. Subject 1531 seems to thoroughly understand the market features in period 1, and then performs the optimum strategy seamlessly in period 2.

  16. 16.

    This chapter and [24] are both under the umbrella of a 3-year NSC research project. Hence, they both share some similar features. What distinguishes [24] from this chapter is that the former explicitly constructs traders’ learning paths in a numerical landscape. The question is then to address whether the observed learning behavior of traders can also be understood as an output of a numerical search algorithm. In other words, they inquire whether there is a connection between behavioral search and numerical search. However, the trading environment here makes it hard to derive this geometrical representation; therefore, the use of a learning index becomes another way to see how this trap might actually also exist. Despite this difference, the implication of these two studies is the same: we need to equip artificial agents with different CI tools so that their search behavior can be meaningfully connected to the cognitive capacity of human traders, or, more directly, we need to reflect upon the cognitive capacity of different CI tools [19].

  17. 17.

    As the psychological literature points out, high intelligence does not always contribute to high performance—the significance of intelligence in performance is more salient when the problems are more complex [25]. In addition, it appears that intelligence exhibits a decreasing marginal contribution in terms of performances [29, 37]. In the setting of an agent-based double auction market, Chen et al. [23] have replicated this diminishing marginal contribution of cognitive capacity. In that article, autonomous traders are modeled by genetic programming with different population size. The population size is manipulated as a proxy variable for working memory capacity. They then found that, while the trading performance between agents with small population size and agents with a large one is significantly different, this difference between agents with a large one and agents with a larger one is negligible.

  18. 18.

    There is no official translation of the seven factors. An attempt to do so on our own is not easy, in particular if one wants to describe the whole of 15–20 adjectives using a single word, such as conscientiousness. What we do here is to follow OCEAN closely and to use the same name if the factor in Big Seven shares very much in common with one of the Big Five. Examples are conscientiousness, agreeableness, and extraversion.

  19. 19.

    Conscientiousness is found to be a good predictor of job performance, mortality, divorce, educational attainment, car accidents, and credit score [1, 7, 8, 24, 38, 49].

  20. 20.

    In fact, both our student t-test and Wilcoxon rank sum test only find this factor statistically significantly different between the performing group and the non-performing group.

  21. 21.

    However, this study differs from [24] in using a different performance measure. See also footnote 16.

  22. 22.

    Among the seven factors, Chen et al. [24] find conscientiousness, extraversion, and agreeableness to be influential, at least in some contexts. In their analysis, they attempt to justify each of these three. Among the three, conscientiousness is probably the easiest one to justify, given the already lengthily archived documents (see footnote 2.4.3). They then go further to justify the other two by using [28] to argue that extraverted subjects are more sensitive to potential rewarding stimuli through the mesolimbic dopamine, which may in turn help them more easily find the more profitable trading arrangements. In addition, for agreeableness, they argue that subjects with a higher degree of agreeableness can resist time pressure and may be able to think for a longer time before making decisions.

References

  1. J. Anderson, S. Burks, C. DeYoung, A. Rustichinid, Toward the Integration of Personality Theory and Decision Theory in the Explanation of Economic Behavior (University of Minnesota, Mimeo, 2011)

    Google Scholar 

  2. J. Arifovic, Genetic algorithms learning and the cobweb model. J. Econ. Dyn. Contr. 18(1), 3–28 (1994)

    Google Scholar 

  3. J. Arifovic, R. McKelvey, S. Pevnitskaya, An initial implementation of the Turing tournament to learning in repeated two-person games. Game. Econ. Behav. 57(1), 93–122 (2006)

    Google Scholar 

  4. B. Arthur, On designing economic agents that behave like human agents. J. Evol. Econ. 3(1), 1–22 (1993)

    Google Scholar 

  5. H. Baker, J. Nofsinger, Behavioral Finance: Investors, Corporations, and Markets (Wiley, New York, 2010)

    Google Scholar 

  6. B. Barber, T. Odean, Trading is hazardous to your wealth: The common stock investment performance of individual investors. J. Finance 55(2), 773–806 (2000)

    Google Scholar 

  7. M. Barrick, M. Mount, The big five personality dimensions and job performance: A meta-analysis. Person. Psychol. 44(1), 1–26 (1991)

    Google Scholar 

  8. M. Barrick, M. Mount, T. Judge, Personality and performance at the beginning of the new millennium: What do we know and where do we go next? Int. J. Sel. Assess. 9, 9–30 (2001)

    Google Scholar 

  9. A. Ben-Ner, F. Halldorsson, Measuring Trust: Which Measures Can be Trusted? Working paper, University of Minnesota (2007)

    Google Scholar 

  10. A. Ben-Ner, L. Putterman, F. Kong, D. Magan, Reciprocity in a two-part dictator game. J. Econ. Behav. Organ. 53(3), 333–352 (2004)

    Google Scholar 

  11. A. Ben-Ner, F. Kong, L. Putterman, Share and share alike? Gender-paring, personality, and cognitive ability as determinants of giving. J. Econ. Psychol. 25, 581–589 (2004)

    Google Scholar 

  12. Y. Bereby-Meyer, A. Roth, The speed of learning in noisy games: Partial reinforcement and the sustainability of cooperation. Am. Econ. Rev. 96(4), 1029–1042 (2006)

    Google Scholar 

  13. Z. Boender, J. Ultee, S. Hovius, Cognitive capacity: No association with recovery of sensibility by Semmes Weinstein test score after peripheral nerve injury of the forearm. J. Plast. Reconstr. Aesthetic Surg. 63(2), 354–359 (2010)

    Google Scholar 

  14. L. Borghans, A. Duckworth, J. Heckman, B. Weel, The Economics and Psychology of Personality Traits, NBER Working Paper No. 13810 (2008)

    Google Scholar 

  15. P. Bossaerts, What decision neuroscience teaches us about financial decision making. Annu. Rev. Financ. Econ. 1, 383–404 (2009)

    Google Scholar 

  16. S.-H. Chen, Computational intelligence in agent-based computational economics, in Computational Intelligence: A Compendium, ed. by J. Fulcher, L. Jain (Springer, New York, 2008), pp. 517–594

    Google Scholar 

  17. S.-H. Chen, Genetic programming and agent-based computational economics: from autonomous agents to product innovation, in Agent-Based Approaches in Economic and Social Complex Systems, ed. by T. Terano, H. Kita, S. Takahashi, H. Deguchi (Springer, New York, 2008), pp. 3–14

    Google Scholar 

  18. S.-H. Chen, Software-agent designs in economics: An interdisciplinary framework. IEEE Comput. Intell. Mag. 3(4), 18–22 (2008)

    Google Scholar 

  19. S.-H. Chen, Collaborative computational intelligence in economics, in Computational Intelligence: Collaboration, Fusion and Emergence, ed. by C. Mumford, L. Jain (Springer, New York, 2009), pp. 233–273

    Google Scholar 

  20. S.-H. Chen, Y.-L. Hsieh, Reinforcement learning in experimental asset markets. E. Econ. J. 37(1), 109–133 (2011)

    Google Scholar 

  21. S.-H. Chen, S. Wang, Emergent complexity in agent-based computational economics. J. Econ. Surv. 25(3), 527–546 (2011)

    Google Scholar 

  22. S.-H. Chen, R.-J. Zeng, T. Yu, Co-evolving trading strategies to analyze bounded rationality in double auction markets, in Genetic Programming Theory and Practice VI, ed. by R. Riolo (Springer, New York, 2008), pp. 195–213

    Google Scholar 

  23. S.-H. Chen, C.-C. Tai, S. Wang, Does cognitive capacity matter when learning using genetic programming in double auction markets? in Multi-Agent-Based Simulation, ed. by G. Di Tosto, H. Van Dyke Parunak. Lecture Notes in Artificial Intelligence, vol. 5683 (Springer, New York, 2010), pp. 37–48

    Google Scholar 

  24. S.-H. Chen, U. Gostoli, C.-C. Tai, K.-C. Shih, To whom and where the hill becomes difficult to climb: Effects of personality and cognitive capacity in experimental DA markets. Adv. Behav. Finance Econ (forthcoming)

    Google Scholar 

  25. R. Christal, W. Tirre, P. Kyllonen, Two for the money: speed and level scores from a computerized vocabulary test, in Proceedings of the Ninth Annual Symposium, Psychology in the Department of Defense (USAFA TR ’84-2), ed. by G. Lee, Ulrich (U.S. Air Force Academy, Colorado Springs, 1984)

    Google Scholar 

  26. A. Conway, M. Kane, R. Engle. Working memory capacity and its relation to general intelligence. Trends Cognit. Sci. 7(12), 547–552 (2003)

    Google Scholar 

  27. M. Daneman, P. Carpenter, Individual differences in integrating information between and within sentences. J. Exp. Psychol. Learn. Mem. Cognit. 9, 561–584 (1983)

    Google Scholar 

  28. R. Depue, P. Collins, Neurobiology of the structure of personality: Dopamine, facilitation of incentive motivation, and extraversion. Behav. Brain Sci. 22(3), 491–517 (1999)

    Google Scholar 

  29. D. Detterman, M. Daniel, Correlations of mental tests with each other and with cognitive variables are highest for low-IQ groups. Intelligence 13, 349–359 (1989)

    Google Scholar 

  30. M. Fenton-O’Creevy, E. Soane, N. Nicholson, P. Willman, Thinking, feeling and deciding: The influence of emotions on the decision making and performance of traders. J. Organ. Behav. 32(8), 1044–1061 (2011)

    Google Scholar 

  31. M. Gendreau, J.-Y. Potvin, Handbook of Metaheuristics (Springer, New York, 2010)

    Google Scholar 

  32. G. Gigerenzer, Gut Feelings: The Intelligence of the Unconscious (Penguin, New York, 2007)

    Google Scholar 

  33. G. Gigerenzer, C. Engel, Heuristics and the Law (MIT, Cambridge, 2006)

    Google Scholar 

  34. T. Gilovich, D. Griffin, D. Kahneman, Heuristics and Biases: The Psychology of Intuitive Judgment (Cambridge University Press, Cambridge, 2002)

    Google Scholar 

  35. J. Hirsh, J. Peterson, Extraversion, neuroticism, and the prisoners dilemma. Pers. Indiv. Differ. 46(2), 254–256 (2009)

    Google Scholar 

  36. M. Hsu, M. Bhatt, R. Adolphs, D. Tranel, C. Camerer, Neural systems responding to degrees of uncertainty in human decision-making. Science 310, 1680–1683 (2005)

    Google Scholar 

  37. E. Hunt, The role of intelligence in modern society. Am. Sci. July/August, 356–368 (1995)

    Google Scholar 

  38. T. Judge, C. Higgins, C. Thoresen, M. Barrick, The big five personality traits, general mental ability, and career success across the life span. Person. Psychol. 52, 621–652 (1999)

    Google Scholar 

  39. M. Kane, D. Hambrick, A. Conway, Working memory capacity and fluid intelligence are strongly related constructs: Comment on Ackerman, Beier, and Boyle. Psychol. Bull. 131, 66–71 (2005)

    Google Scholar 

  40. R. Kaplan, D. Norton, The Balanced Scorecard: Translating Strategy into Action (Harvard Business Press, Boston, 1996)

    Google Scholar 

  41. P. Kyllonen, R. Christal, Reasoning ability is (little more than) working-memory capacity? Intelligence 3, 1–64 (1990)

    Google Scholar 

  42. S. Lewandowsky, K. Oberauer, L.-X. Yang, U. Ecker, A working memory test battery for MatLab. Behav. Res. Meth. 42(2), 571–585 (2011)

    Google Scholar 

  43. A. Lo, D. Repin, The psychophysiology of real-time financial risk processing. J. Cognit. Neurosci. 14(3), 323–339 (2002)

    Google Scholar 

  44. A. Lo, D. Repin, B. Steenbarger, Fear and greed in financial markets: A clinical study of day-traders. Am. Econ. Rev. 95(2), 352–359 (2005)

    Google Scholar 

  45. B. Lucey, M. Dowling, The role of feelings in investor decision-making. J. Econ. Surv. 19(2), 211–237 (2005)

    Google Scholar 

  46. K. Morsanyi, S. Handley, How smart do you need to be to get it wrong? The role of cognitive capacity in the development of heuristic-based judgment. J. Exp. Child Psychol. 99, 18–36 (2008)

    Google Scholar 

  47. K. Oberauer, H.-M. Süß, O. Wilhelm, W. Wittman, The multiple faces of working memory: Storage, processing, supervision, and coordination. Intelligence 31, 167–193 (2003)

    Google Scholar 

  48. T. Offerman, J. Sonnemans, What’s causing overreaction? An experimental investigation of recency and the hot-hand effect. Scand. J. Econ. 106(3), 533–554 (2004)

    Google Scholar 

  49. B. Roberts, N. Kuncel, R. Shiner, A. Caspi, L. Goldberg, The power of personality: The comparative validity of personality traits, socio-economic status, and cognitive ability for predicting important life outcomes. Perspect. Psychol. Sci. 2, 313–345 (2007)

    Google Scholar 

  50. J. Rust, J. Miller, R. Palmer, Behavior of trading automata in a computerized double auction market, in Double Auction Markets: Theory, Institutions, and Laboratory Evidence, ed. by D. Friedman, J. Rust (Addison Wesley, CA, 1993), pp. 155–198

    Google Scholar 

  51. J. Rust, J. Miller, R. Palmer, Characterizing effective trading strategies: Insights from a computerized double auction tournament. J. Econ. Dyn. Contr. 18, 61–96 (1994)

    Google Scholar 

  52. H. Shefrin, Beyond Greed and Fear: Understanding Behavioral Finance and the Psychology of Investing (Oxford University Press, New York, 2007)

    Google Scholar 

  53. V. Smith, Papers in Experimental Economics (Cambridge University Press, Cambridge, 1991)

    Google Scholar 

  54. E. Weber, E. Johnson, Decisions under uncertainty: Psychological, economic, and neuroeconomic explanations of risk preference, in Neuroeconomics: Decision Making and the Brain, ed. by P. Glimcher, E. Fehr, A. Rangel, C. Camerer, R. Poldrack (Academic, New York, 2008), pp. 127–144

    Google Scholar 

  55. M. Wellman, A. Greenwald, P. Stone, Autonomous Bidding Agents: Strategies and Lessons from the Trading Agent Competition (MIT, Cambridge, 2007)

    Google Scholar 

  56. M. **a, J. Stallaert, A.B. Whinston, Solving the combinatorial double auction problem. Eur. J. Oper. Res. 164, 239–251 (2005)

    Google Scholar 

Download references

Acknowledgements

The authors are grateful to Professor Kuo-Shu Yang for his generous permission for using his developed Chinese version of the Big-Five personality test. We are also grateful to Professor Li-Jen Weng and Professor Lei-**eng Yang for their advice and guidance on the psychological tests implemented in this study. NSC research grants no. 98-2410-H-004-045-MY3, no. 99-2811-H-004-014, and no. 100-2410-H-029-001 are also gratefully acknowledged.

Author information

Authors and Affiliations

  1. AIECON Research Center, Department of Economics, National Chengchi University, Chengchi, Taiwan

    Shu-Heng Chen & Kuo-Chuan Shih

  2. Department of Economics, Tunghai University, Tunghai, Taiwan

    Chung-Ching Tai

Authors
  1. Shu-Heng Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Kuo-Chuan Shih
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Chung-Ching Tai
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shu-Heng Chen .

Editor information

Editors and Affiliations

  1. Dept. Production Engineering, and Management, Technical University of Crete, University Campus, Chania, 731 00, Greece

    Michael Doumpos

  2. , Department of Production Engineering and, Technical University of Crete, University Campus, Chania, 73100, Greece

    Constantin Zopounidis

  3. , Department of Industrial and Systems Eng, University of Florida, Weil Hall 401, Gainesville, 32611, Florida, USA

    Panos M. Pardalos

Rights and permissions

Reprints and permissions

Copyright information

© 2012 Springer Science+Business Media New York

About this chapter

Cite this chapter

Chen, SH., Shih, KC., Tai, CC. (2012). Can Artificial Traders Learn and Err Like Human Traders? A New Direction for Computational Intelligence in Behavioral Finance. In: Doumpos, M., Zopounidis, C., Pardalos, P. (eds) Financial Decision Making Using Computational Intelligence. Springer Optimization and Its Applications, vol 70. Springer, Boston, MA. https://doi.org/10.1007/978-1-4614-3773-4_2

Download citation

Publish with us

Policies and ethics

Search

Navigation