Agent-based Artificial Stock Market Research Articles

Noise traders in an agent-based artificial stock market

AbstractThis paper investigates whether noise traders can survive in the long run and how they influence financial markets by proposing an agent-based artificial stock market, as one simulation model of computational economics. This market contains noise traders, informed and uninformed traders. Informed and uninformed traders can learn from information by using Genetic Programming, while noise traders cannot. The system is first calibrated to real financial markets by replicating several stylized facts. We find that noise traders cannot survive or they just transform to other kind of traders in the long run, and they increase market volatility, price distortion, noise trader risk, and trading volume in the market. However, regulation intervention, e.g., price limits, transaction tax and longer settlement cycle, can affect noise trader’s surviving period and their influence on markets.

Impacts of social networks in an agent-based artificial stock market

ABSTRACT We propose an agent-based artificial stock market to investigate the influences of social networks on financial markets. It contains four types of traders whose information sets and trading strategies are different. The application of artificial intelligence is employed in informed and uninformed traders’behaviour and heterogeneity. When information is exogenous, social networks result in higher volatility and trading volume and lower price distortion and bid-ask spread. When information is endogenous, the influences are reversed. The reason is that social networks harm information production after traders tend to rely on information from communication, instead of spending a cost on it..

Quantifying the concerns of Dimon and Buffett with data and computation

Despite considerable efforts, the determinants of firm growth and financial market volatility have not been definitively identified, yet sets of stylized facts – most notably power-law distributions – relating to firm size and market returns suggest both evolve as part of a complex system. This scenario implies that a positive feedback loop between firms and investors exists and may be responsible for prejudicing the way management allocates their resources, with firm, and economic, growth adversely affected. Further, there are growing real-world concerns that the management of publicly listed firms is becoming too concerned with the movement of their firm's share price, which is adversely influencing resource allocation decisions. A related concern is that agents, in general, within financial markets are placing a disproportionate focus on short-term factors. To investigate the ramifications of the proposed feedback loop on firm growth and market volatility, this paper implements a novel agent-based artificial stock market where management can consider the movements of their firm's endogenously determined share price when allocating resources between sales and margin growth. The results highlight an inferior outcome regarding firm growth, and various other financial metrics, if management is overly concerned with share price movements. The growth of the firms (and market) is also affected by the mixture of the investor classes initiated due to the divergent levels of volatility they create. Additionally, the model presents insights into how and why the extent to which agents consider past outcomes in their decision-making process becomes influential. Notably, the model's results emulate an extensive set of global micro-level firm data. By providing significant insights on the effects of the stock market on management decision-making and its ramifications for firm growth, this paper provides crucial insights into the mechanisms responsible for inefficient behavior by market participants.

Effects of fundamentals acquisition and strategy switch on stock price dynamics

An agent-based artificial stock market is developed to simulate trading behavior of investors. In the market, acquisition and employment of information about fundamentals and strategy switch are investigated to explain stock price dynamics. Investors could obtain the information from both market and neighbors resided on their social networks. Depending on information status and performances of different strategies, an informed investor may switch to the strategy of fundamentalist. This in turn affects the information acquisition process, since fundamentalists are more inclined to search and spread the information than chartists. Further investigation into price dynamics generated from three typical networks, i.e. regular lattice, small-world network and random graph, are conducted after general relation between network structures and price dynamics is revealed. In each network, integrated effects of different combinations of information efficiency and switch intensity are investigated. Results have shown that, along with increasing switch intensity, market and social information efficiency play different roles in the formation of price distortion, standard deviation and kurtosis of returns.

Evolutionary Frequency and Forecasting Accuracy: Simulations Based on an Agent-Based Artificial Stock Market

Learning how to forecast is always important for traders, and divergent learning frequencies prevail among traders. The influence of the evolutionary frequency on learning performance has occasioned many studies of agent-based computational finance (e.g., Lettau in J Econ Dyn Control 21:1117–1147, 1997. doi: 10.1016/S0165-1889(97)00046-8 ; Szpiro in Complexity 2(4):31–39, 1997. doi: 10.1002/(SICI)1099-0526(199703/04)2:4 3.0.CO;2-3 ; Cacho and Simmons in Aust J Agric Resour Econ 43(3):305–322, 1999. doi: 10.1111/1467-8489.00081 ). Although these studies all suggest that evolving less frequently and, hence, experiencing more realizations help learning, this implication may result from their common stationary assumption. Therefore, we first attempt to approach this issue in a ‘dynamically’ evolving market in which agents learn to forecast endogenously generated asset prices. Moreover, in these studies’ market settings, evolving less frequently also meant having a longer time horizon. However, it is not true in many market settings that are even closer to the real financial markets. The clarification that the evolutionary frequency and the time horizon are two separate notions leaves the effect of the evolutionary frequency on learning even more elusive and worthy of exploration independently. We find that the influence of a trader’s evolutionary frequency on his forecasting accuracy depends on all market participants and the resulting price dynamics. In addition, prior studies also commonly assume that traders have identical preferences, which is too strong an assumption to apply to a real market. Considering the heterogeneity of preferences, we find that converging to the rational expectations equilibrium is hardly possible, and we even suggest that agents in a slow-learning market learn frequently. We also apply a series of econometric tests to explain the simulation results.

A comparison of U.S and Chinese financial market microstructure: heterogeneous agent-based multi-asset artificial stock markets approach

The market microstructure literatures study how the traders work in the financial market. In this paper, we propose a novel heterogeneous agent-based multi-asset artificial stock market based on Santa Fe Artificial Stock Market (SFI-ASM) to compare the financial market microstructure between U.S. and China. We first develop a set of new parameters for the single stock market simulation to improve the way that agents monitor the market and choose different strategies, which make our model closer to the real financial market. Secondly, we construct a multiple assets financial market by incorporating two new types of agents, namely, zero-intelligence agents and less-intelligence agents, and conduct simulations for different evolution speeds, strategies, and intelligence levels to achieve the optimal models of Chinese and U.S. financial markets before and after the financial crisis. Based on the simulation results, we present a comprehensive analysis of the market microstructure for the two financial markets.

Exploring issues of market inefficiency by the role of forecasting accuracy in survivability

Whether markets are efficient or not has been broadly discussed in the empirical literature since the efficient markets hypothesis was proposed by Fama and others in the 1960s. Unfortunately, they did not come to a consistent conclusion. Besides, while these studies show whether a specific market is efficient or not, little has been done to explore the issues regarding the degree of market inefficiency. This paper attempts to resolve the puzzle of the inconsistent conclusions in the empirical literature by adopting a bottom-up approach which takes market participants’ interactions and coordination into consideration. By simulating an agent-based artificial stock market, this paper concludes with three main findings. First, agents’ survivability is mainly decided by risk preference, and not forecasting accuracy. Survivors may have diverse forecasting accuracy. Second, because market prices are not decided by agents based on accurate predictions, markets can not be efficient. What may exist is only the difference of the degree of inefficiency between markets. Third, the more relevant to survivability the forecasting accuracy in a market is, the less inefficient the market will be. Therefore, this paper suggests that it may be better to view the divergent empirical results regarding market efficiency as a fact that markets are inefficient to a variety of degrees.

Information identification in different networks with heterogeneous information sources

Traditional cheap-talk game model with homogeneous information sources provided a conclusion that dishonest information sources will not be identified if he changes strategy stochastically. In this paper, the authors incorporate different information diffusion networks and heterogeneous information sources into an agent-based artificial stock market. The obtained results are different with traditional results that identification ability of uninformed agents has been highly improved with diffusion networks and heterogeneous information sources. Additionally, the authors find uninformed agents can improve identification ability only if there exists a sufficient number of heterogeneous information sources in stock market.

MICROSTRUCTURE DYNAMICS AND AGENT-BASED FINANCIAL MARKETS: CAN DINOSAURS RETURN?

This paper formalizes observations made under agent-based artificial stock market models into a concrete hypothesis, which is called the Dinosaur Hypothesis. This hypothesis states that the behavior of financial markets constantly changes and that the trading strategies in a market need to continuously co-evolve with it in order to remain effective. After formalizing the hypothesis, we suggest a testing methodology and run tests under 10 international financial markets. Our tests are based on a framework that we recently developed, which used Genetic Programming as a rule inference engine, and Self-Organizing Maps as a clustering machine for the above rules. However, an important assumption of that study was that maps among different periods were directly comparable with each other. In order to allow this to happen, we had to keep the same clusters throughout the different time periods of our experiments. Nevertheless, this assumption could be considered as strict or even unrealistic. In this paper, we relax this assumption. This makes our model more realistic. In addition, this allows us to investigate in depth the dynamics of market behavior and test for the plausibility of the Dinosaur Hypothesis. The results show that indeed markets' behavior constantly changes. As a consequence, strategies need to continuously co-evolve with the market; if they do not, they become obsolete or dinosaurs.

Intelligent agent-assisted adaptive order simulation system in the artificial stock market

Agent-based computational economics (ACE) has received increased attention and importance over recent years. Some researchers have attempted to develop an agent-based model of the stock market to investigate the behavior of investors and provide decision support for innovation of trading mechanisms. However, challenges remain regarding the design and implementation of such a model, due to the complexity of investors, financial information, policies, and so on. This paper will describe a novel architecture to model the stock market by utilizing stock agent, finance agent and investor agent. Each type of investor agent has a different investment strategy and learning method. A prototype system for supporting stock market simulation and evolution is also presented to demonstrate the practicality and feasibility of the proposed intelligent agent-based artificial stock market system architecture.

Liquidity cost of market orders in the Taiwan Stock Market: A study based on an order-driven agent-based artificial stock market

We developed an order-driven agent-based artificial stock market to analyze the liquidity costs of market orders in the Taiwan Stock Market (TWSE). The agent-based stock market was based on the DFGIS model proposed by Daniels, Farmer, Gillemot, Iori and Smith (Daniels et al., 2003). We also improved the DFGIS model by using two average order size parameters. When tested on 10 stocks and securities in the market, the model-simulated liquidity costs were higher than those of the TWSE data. We identified some possible factors that have contributed to this result: 1) the overestimated effective market order size, which can be improved by using two average order size parameters; 2) the random market order arrival time designed in the DFGIS model; 3) the zero-intelligence of the artificial agents in our model; and 4) the price of the effective market order. We continued improving the model so that it could be used to study liquidity costs and to devise liquidation strategies for stocks and securities traded in the Taiwan Stock Market.

Dynamic trade execution: a grammatical evolution approach

Trade execution is concerned with the actual mechanics of buying or selling the desired amount of a financial instrument. Investors wishing to execute large orders face a tradeoff between market impact and opportunity cost. Trade execution strategies are designed to balance out these costs, thereby minimising total trading cost. Despite the importance of optimising the trade execution process, this is difficult to do in practice due to the dynamic nature of markets and due to our imperfect understanding of them. In this paper, we adopt a novel approach, combining an evolutionary methodology whereby we evolve high-quality trade execution strategies, with an agent-based artificial stock market, wherein the evolved strategies are tested. The evolved strategies are found to outperform a series of benchmark strategies and several avenues are suggested for future work.

On the plausibility of sunspot equilibria

This study examines the plausibility of the emergence of sunspot equilibria in an agent-based artificial stock market. Using the agent-based model, we make the sunspots explicit so that we can test, e.g., by means of the Granger causality test, whether purely extrinsic uncertainty can influence price dynamics. In addition, through agent-based simulation, the coordination process, which is mainly driven by genetic programming, becomes observable, which enables us to analyze what agents perceive and whether they believe in sunspots. By manipulating different control variables, three series of experiments are conducted. Generally speaking, the chances of observing “sunspot equilibria” in this agent-based artificial stock market are small. However, the sunspot believers can never be driven out of the market. Nevertheless, they are always outnumbered by fundamental believers, which is evidence that the market as collective behavior is rational. We also find that lengthening the time horizon will make it difficult for sunspot believers to survive.

Risk preference, forecasting accuracy and survival dynamics: Simulations based on a multi-asset agent-based artificial stock market

The relevance of risk preference and forecasting accuracy to the survival of investors is an issue that has recently attracted a number of theoretical studies. By using agent-based computational modeling, this paper extends the existing studies to an economy where adaptive behaviors are autonomous and complex heterogeneous. Specifically, a computational multi-asset artificial stock market corresponding to Blume and Easley [Blume, L., Easley, D., 1992. Evolution and market behavior. Journal of Economic Theory 58, 9–40] and Sandroni [Sandroni, A., 2000. Do markets favor agents able to make accurate predictions? Econometrica 68, 1303–1341] is constructed and studied. Through simulation, we present results that contradict the market selection hypothesis.

Relative risk aversion and wealth dynamics

As a follow-up to the work of Chen and Huang [S.-H. Chen, Y.-C. Huang, Risk preference, forecasting accuracy and survival dynamics: simulations based on a multi-asset agent-based artificial stock market, Working Paper Series 2004-1, AI-ECON Research Center, National Chengchi University, 2004; S.-H. Chen, Y.-C. Huang, Risk preference and survival dynamics, in: T. Terano, H. Kita, T. Kaneda, K. Arai, H. Deghchi (Eds.), Agent-Based Simulation: From Modeling Methodologies to Real-World Applications, Springer Series on Agent-Based Social Systems, vol. 1, 2005, pp. 135–143], this paper continues to explore the relationship between wealth share dynamics and risk preferences in the context of an agent-based multi-asset artificial stock market. We simulate a multi-asset agent-based artificial stock market composed of heterogeneous agents with different degrees of relative risk aversion. As before, we find that the difference in risk aversion and the resultant saving behavior are the primary forces in determining the survivability of agents. In addition to the stability of the saving behavior, the level of the saving rate also plays a crucial role. The agents with stable saving behavior, e.g., the log-utility agents, may still become extinct because of their low saving rates, whereas the agents with unstable saving behavior may survive because of their high saving rates, implied by their highly risk-averse preferences.

Simulation of game analysis based on an agent-based artificial stock market re-examined

This work re-examined the simulation result of game analysis (Joshi et al., 2000) based on an agent-based model, Santa Fe Institute Artificial Stock Market. Allowing for recent research work on this artificial model, this paper’s modified game simulations found that the dividend amplitude parameter is a crucial factor and that the original conclusion still holds in a not long period, but only when the dividend amplitude is large enough. Our explanation of this result is that the dividend amplitude parameter is a measurement of market uncertainty. The greater the uncertainty, the greater the price volatility, and so is the risk of investing in the stock market. The greater the risk, the greater the advantage of including technical rules.

On the emergent properties of artificial stock markets: the efficient market hypothesis and the rational expectations hypothesis

By studying two well known hypotheses in economics, this paper illustrates how emergent properties can be shown in an agent-based artificial stock market. The two hypotheses considered are the efficient market hypothesis and the rational expectations hypothesis. We inquire whether the macrobehavior depicted by these two hypotheses is consistent with our understanding of the microbehavior. In this agent-based model, genetic programming is applied to evolving a population of traders learning over time. We first apply a series of econometric tests to show that the EMH and the REH can be satisfied with some portions of the artificial time series. Then, by analyzing traders’ behavior, we show that these aggregate results cannot be interpreted as a simple scaling-up of individual behavior. A conjecture based on sunspot-like signals is proposed to explain why macrobehavior can be very different from microbehavior. We assert that the huge search space attributable to genetic programming can induce sunspot-like signals, and we use simulated evolved complexity of forecasting rules and Granger causality tests to examine this assertion.

Evolving traders and the business school with genetic programming: A new architecture of the agent-based artificial stock market

In this paper, we propose a new architecture to study artificial stock markets. This architecture rests on a mechanism called ‘school’ which is a procedure to map the phenotype to the genotype or, in plain English, to uncover the secret of success. We propose an agent-based model of ‘ school’, and consider school as an evolving population driven by single-population GP (SGP). The architecture also takes into consideration traders’ search behavior. By simulated annealing, traders’ search density can be connected to psychological factors, such as peer pressure or economic factors such as the standard of living. This market architecture was then implemented in a standard artificial stock market. Our econometric study of the resultant artificial time series evidences that the return series is independently and identically distributed (iid), and hence supports the efficient market hypothesis (EMH). What is interesting though is that this iid series was generated by traders, who do not believe in the EMH at all. In fact, our study indicates that many of our traders were able to find useful signals quite often from business school, even though these signals were short-lived.