THE IMPLICATIONS OF TRADER COGNITIVE ABILITIES ON STOCK MARKET PROPERTIES

• Author: Viktor Manahov,Mona Soufian,Robert Hudson
• Published date: 01 January 2014
• Date: 01 January 2014
• DOI: http://doi.org/10.1002/isaf.1348

THE IMPLICATIONS OF TRADER COGNITIVE ABILITIES ON

STOCK MARKET PROPERTIES

VIKTOR MANAHOV,

a

*MONA SOUFIAN

b

AND ROBERT HUDSON

c

a

Newcastle University Business School, Newcastle University, Newcastle upon Tyne, UK

b

Newcastle Business School, Northumbria University, Newcastle upon Tyne,UK

c

University of Hull, Yorkshire,UK

SUMMARY

Market structure and individual rationality remain at the centre of a debate as to which is the main driving force in

market performance. We examine the role of individual rationality, comparing zero-intelligence traders with traders

with different levels of intelligence using a special adaptive form of strongly typed genetic programming-based

learning algorithm. We use this approach with real data: historical quotes of the S&P 500 and Coca-Cola stock

prices. We find a mixture of positive and negative impacts from intelligence on market performance. Because

the concept of market structure as a driving force has been significantly challenged in the literature, we suggest that

the inclusion of both intelligence and market structures is important when examining the driving forces of market

performance.This inclusion is consistentwith the research of Toddand Gigerenzer (Journalof Economic Psychology,

24 (2003) 143–165), which asserts that both environment (market structure) and agents’cognition play important

roles. Copyright © 2013 John Wiley & Sons,Ltd.

Keywords: agent-based modelling; artificial stock market; genetic programming; zero intelligence; stylized facts;

efficient market hypothesis

1. INTRODUCTION

The empirical results regarding the extent to which individuals’intelligence or market structure influence

market performance and market efficiency have been controversial. There is no consensus as to whether

market performance and efficiency can be improved through the structural (institutional rulesand regula-

tory) aspects of the market or whether it is the intelligence of the traders that matters. Some research

emphasizes the importance of the intelligence of traders in market performance (Yeh, 2007, 2008). This

perspective is contradictory to other research that claims that market structure is the main driving force

of efficiency, in that zero-intelligence (ZI) agents have the same impact as intelligent agents on market

performance and, thus, the intelligence of the agents has no significant impact on efficiency relative to

market structure (Gode and Sunder, 1993; Chen and Tai, 2003).

The reason behind this divergence in the literature is that the essence of individual cognition and its

interaction with the market have not received the attention of financial scholars. Because environments

are constructed and shaped by the cognitions of the decision makers who act in them, one can conclude

that the emergent structure of the market corresponds to the structure of the individuals’cognitions.

Subsequently, the emergent structure of the market interacts with the cognitions of the individuals. In

her heterogeneous learning experiment with agents’limited rationality, Giannitsarou (2003) argued that

* Correspondence to: Viktor Manahov, Newcastle University Business School, 6th Floor, PhD Office 6.22, 5 Barrack Road,

Newcastle upon Tyne, NE1 4SE, UK. E-mail: v.manahov@newcastle.ac.uk

Copyright © 2013 John Wiley & Sons, Ltd.

INTELLIGENT SYSTEMS IN ACCOUNTING, FINANCE AND MANAGEMENT

Intell. Sys. Acc. Fin. Mgmt. 21, 1–18 (2014)

Published online 12 November 2013 in Wiley OnlineLibrary (wileyonlinelibrary.com) DOI: 10.1002/isaf.1348

the representative agent is often a good approximation of the agents in an economy. Therefore, both in-

telligence and market structure must be included when examining market performance. Todd and

Gigerenzer (2003), in their challenge to the perfect rationality of individuals, stressed the importance

of including both the environment and the agents’cognition. In their discussion, they introduced

ecological rationality which builds on the perspective of Simon’s bounded rationality. Applying the

essence of ecological rationality can provide us with a better understanding of how both the environment

and the agents’cognition are determinants and driving forces of market performance and can help us to

understand the reasons for the divergence in the results when we include only one aspect, intelligence

or environmental structure.

Widiputra et al. (2009) described the behaviour of multiple stock markets within the framework of a

dynamic interaction network (DIN) capable of developing various dynamic interactions between genes

and predicting their future expressions. The DIN model revealed complex dynamic relationships

between stock markets that went beyond the scope of traditional econometric models. In this paper,

we follow Yeh (2007), who supports the importance of intelligence as a driving force in market

performance, albeit that he recommends the interpretation of the results with certain caveats because

he applied the artificial data framework of the Santa Fe Artificial Market (SF-ASM).

Using real historical daily closing prices of the S&P 500 and stocks in the Coca-Cola Company and a

different computational technique, this paper provides a new perspective on the research in this field by

explaining the reason for the divergence of the results in the literature, as neither intelligence nor

market structure individually dominates in driving market performance. As expected, the results for

our tests, using real data, are different from the results of Yeh (2007, 2008), who used simulated market

SF-ASM and fixed arti ficial data. We obtain a mixture of positive and negative impacts from individual

intelligence on market performance. Our empirical results indicate that using only individual intelli-

gence provides us with an incomplete picture. Consistent with Todd and Gigerenzer (2003), we suggest

that both intelligence and market structure are equally important and consistent with Yeh (2007, 2008);

we suggest that further research should include both intelligence and market structure.

The structure of this paper is organized as follows. Section 2 presents the background and reviews the

literature in the field. Section 3 discusses the experimental design, and Section 4 discusses the simulation

results. The paperconcludes with overall remarks and discusses avenues for further researchin Section 5.

2. BACKGROUND

The topics of market structure and individual rationality have been at the centre of research into markets

for many years. Academics have been divided into two main camps. One camp believes that market

structure is the main driving force behind market performance and argues that adequate market rules

are of prime importance and individual rationality plays an insignificant role in the formation of market

properties. The other camp believes that individual rationality is the most important factor in market

performance. These authors emphasize the role of intelligence and contend that market str uctures are

less dominant.

In a seminal paper, Gode and Sunder (1993) investigated the relationship between individual

motivations and market efficiency. The authors implemented ZI traders of a type that ‘does not seek

to maximize profits, and does not observe, remember, or learn’and concluded that allocative

efficiency is not dependent on individual rationality or trader strategy but that market structure is

the main driving force. Sunder (2006a,2006b) suggested a new direction of research that is entirely

concentrated on institutions and structures instead of individual human behaviour. Sunder (2007)

2V. MANAHOV ETAL.

Copyright © 2013 John Wiley & Sons, Ltd. Intell. Sys. Acc. Fin. Mgmt., 21,1–18(2014)

DOI: 10.1002/isaf