Agent-based Computational Economics Research Articles

Fuzzy logic-based bidding strategies in dynamic double auctions

PurposeDouble auctions are widely used market mechanisms on the world. Communication technologies such as internet increased importance of this market institution. The purpose of this study is to develop novel bidding strategies for dynamic double auction markets, explain price formation through interactions of buyers and sellers in decentralized fashion and compare macro market outputs of different micro bidding strategies.Design/methodology/approachIn this study, two novel bidding strategies based on fuzzy logic are presented. Also, four new bidding strategies based on price targeting are introduced for the aim of comparison. The proposed bidding strategies are based on agent-based computational economics approach. The authors performed multi-agent simulations of double auction market for each suggested bidding strategy. For the aim of comparison, the zero intelligence strategy is also used in the simulation study. Various market outputs are obtained from these simulations. These outputs are market efficiencies, price means, price standard deviations, profits of sellers and buyers, transaction quantities, profit dispersions and Smith’s alpha statistics. All outputs are also compared to each other using t-tests and kernel density plots.FindingsThe results show that fuzzy logic-based bidding strategies are superior to price targeting strategies and the zero intelligence strategy. The authors also find that only small number of inputs such as the best bid, the best ask, reference price and trader valuations are sufficient to take right action and to attain higher efficiency in a fuzzy logic-based bidding strategy.Originality/valueThis paper presents novel bidding strategies for dynamic double auction markets. New bidding strategies based on fuzzy logic inference systems are developed, and their superior performances are shown. These strategies can be easily used in market-based control and automated bidding systems.

Connectionist agent-based learning in bank-run decision making.

It is of utter importance for the policy makers, bankers, and investors to thoroughly understand the probability of bank-run (PBR) which was often neglected in the classical models. Bank-run is not merely due to miscoordination (Diamond and Dybvig, 1983) or deterioration of bank assets (Allen and Gale, 1998) but various factors. This paper presents the simulation results of the nonlinear dynamic probabilities of bank runs based on the global games approach, with the distinct assumption that heterogenous agents hold highly correlated but unidentical beliefs about the true payoffs. The specific technique used in the simulation is to let agents have an integrated cognitive-affective network. It is observed that, even when the economy is good, agents are significantly affected by the cognitive-affective network to react to bad news which might lead to bank-run. Both the rise of the late payoffs, R, and the early payoffs, r, will decrease the effect of the affective process. The increased risk sharing might or might not increase PBR, and the increase in late payoff is beneficial for preventing the bank run. This paper is one of the pioneers that links agent-based computational economics and behavioral economics.

A Bright Future for Financial Agent-Based Models

The history of research in finance and economics has been widely impacted by the field of Agent- based Computational Economics (ACE). While at the same time being popular among natural science researchers for its proximity to the successful methods of physics and chemistry for example, the field of ACE has also received critics by a part of the social science community for its lack of empiricism. Yet recent trends have shifted the weights of these general arguments and potentially given ACE a whole new range of realism. At the base of these trends are found two present-day major scientific breakthroughs: the steady shift of psychology towards a hard science due to the advances of neuropsychology, and the progress of artificial intelligence and more specifically machine learning due to increasing computational power and big data. These two have also found common fields of study in the form of computational neuroscience, and human-computer interaction, among others. We outline here the main lines of a computational research study of collective economic behavior via Agent-Based Models (ABM) or Multi-Agent System (MAS), where each agent would be endowed with specific cognitive and behavioral biases known to the field of neuroeconomics, and at the same time autonomously implement rational quantitative financial strategies updated by machine learning. We postulate that such ABMs would offer a whole new range of realism.

The Impact of Quantitative Easing on UK Bank Lending: Why Banks Do Not Lend to Businesses?

The growing proportion of UK bank lending to the financial sector reached a peak in 2007 just before the onset of the Global Financial Crisis (GFC). This marks a trend in the dwindling amount of bank lending to private sector non-financial corporations (PNFCs), which was exacerbated with the Great Recession. Many central banks aimed to revive bank lending with quantitative easing (QE) and unconventional monetary policy. We propose an agent based computational economics (ACE) model which combines the main factors in the economic environment of QE and Basel regulatory framework to analyse why UK banks do not prioritize lending to non-financial businesses. The lower bond yields caused by QE encourage big firms to substitute away from bank borrowing to bond issuance. In addition, the risk weight regime of Basel I/II on capital induces banks to favour mortgages over business loans to small and medium enterprises (SMEs). The combination of lower bond yields and Basel II/III capital requirements on banks, which, respectively, impact demand and supply of credit in the UK, plays a role in the drop of bank loans to businesses. The ACE model aims to reinstate policy regimes that form constraints and incentives for the behaviour of market participants to provide the causal factors in observed macro-economic phenomena.

A hybrid agent-based computational economics and optimization approach for supplier selection problem

Abstract Supplier evaluation and selection problem is among the most important of logistics decisions that have been addressed extensively in supply chain management. The same logistics decision is also important in freight transportation since it identifies trade relationships between business establishments and determines commodity flows between production and consumption points. The commodity flows are then used as input to freight transportation models to determine cargo movements and their characteristics including mode choice and shipment size. Various approaches have been proposed to explore this latter problem in previous studies. Traditionally, potential suppliers are evaluated and selected using only price/cost as the influential criteria and the state-of-practice methods. This paper introduces a hybrid agent-based computational economics and optimization approach for supplier selection. The proposed model combines an agent-based multi-criteria supplier evaluation approach with a multi-objective optimization model to capture both behavioral and economical aspects of the supplier selection process. The model uses a system of ordered response models to determine importance weights of the different criteria in supplier evaluation from a buyers’ point of view. The estimated weights are then used to calculate a utility for each potential supplier in the market and rank them. The calculated utilities are then entered into a mathematical programming model in which best suppliers are selected by maximizing the total accrued utility for all buyers and minimizing total shipping costs while balancing the capacity of potential suppliers to ensure market clearing mechanisms. The proposed model, herein, was implemented under an operational agent-based supply chain and freight transportation framework for the Chicago Metropolitan Area.

Modeling economic systems as locally-constructive sequential games

Real-world economies are open-ended dynamic systems consisting of heterogeneous interacting participants. Human participants are decision-makers who strategically take into account the past actions and potential future actions of other participants. All participants are forced to be locally constructive, meaning their actions at any given time must be based on their local states; and participant actions at any given time affect future local states. Taken together, these essential properties imply real-world economies are locally-constructive sequential games. This paper discusses a modeling approach, Agent-based Computational Economics (ACE), that permits researchers to study economic systems from this point of view. ACE modeling principles and objectives are first concisely presented and explained. The remainder of the paper then highlights challenging issues and edgier explorations that ACE researchers are currently pursuing.

The artificial stock market model based on agent and scale-free network

As a complex giant system, financial market has complexity characteristics with distinctive forms, which shakes theoretical basis of effective market hypothesis to challenge traditional paradigm for financial research. Based on agent-based computational economics, the work established artificial stock market by computer’s high-speed information processing ability and bottom-up modeling method. The real financial markets were highly simulated to guide financial market regulation and policy development. The artificial stock market model in the work realized the simulation of real stock market to a certain extent. This showed that it was effective to improve artificial stock market modeling method based on agent and scale-free network. Therefore, the application value analysis of model has important practical significance.

The Future of Korea's Trade and Business Portfolio in North Africa: A Deep Horizon Political Economy Scan of Algeria, Morocco and Tunisia

Since 2000, Korea's trade volume with North Africa has increased to over $15 billion and a trade surplus of $7 billion. However recent political unrest across North Africa, especially in Algeria, Morocco and Tunisia, could threaten to significantly impact Korea's trade, investment and diplomatic activities across the region. As economists have long established the relationship between political stability, economic growth and foreign direct investment, this study performs a deep horizon scan on domestic political stability in Algeria, Morocco and Tunisia. We use an agent based computational economics (ACE) model called SENTURION to anticipate likely political stability outcomes, scenarios and potential macroeconomic impact. We then offer specific policy and investment advice for Korean business, trade and investment activities in each country, focusing on risk, sectoral and investment strategies given likely political and macroeconomic performance and resiliency in each country.

LINKING AGENT-BASED COMPUTATIONAL ECONOMICS TO POST KEYNESIAN ECONOMICS

Agent-based computational economics is relatively a new methodology in economics. It is defined as ‘the computational modeling of economic processes (including whole economies) as open-ended dynamic systems of interacting agents’. Contrary to fundamental assumptions of neoclassical and mainstream approaches, agent-based computational economics assumes that (a) agents are heterogeneous and bounded rational decision makers in an economy, (b) an economy is a non-linear, complex and adaptive system. Moreover, these assumptions seem to hold true for Post Keynesian economics to a large extent. Given that, this study attempts to bring out similarities and potential links between these two economic thoughts. http://dx.doi.org/10.17130/ijmeb.20173126260

Macroeconomic Policy in DSGE and Agent-Based Models Redux: New Developments and Challenges Ahead

The Great Recession seems to be a natural experiment for economic analysis, in that it has shown the inadequacy of the predominant theoretical framework - the New Neoclassical Synthesis (NNS) - grounded on the DSGE model. In this paper, we present a critical discussion of the theoretical, empirical and political-economy pitfalls of the DSGE-based approach to policy analysis. We suggest that a more fruitful research avenue should escape the strong theoretical requirements of NNS models (e.g., equilibrium, rationality, representative agent, etc.) and consider the economy as a complex evolving system, i.e. as an ecology populated by heterogenous agents, whose far-from-equilibrium interactions continuously change the structure of the system. This is indeed the methodological core of agent-based computational economics (ACE), which is presented in this paper. We also discuss how ACE has been applied to policy analysis issues, and we provide a survey of macroeconomic policy applications (fiscal and monetary policy, bank regulation, labor market structural reforms and climate change interventions). Finally, we conclude by discussing the methodological status of ACE, as well as the problems it raises.

Book Reviews

Leigh Tesfatsion of Iowa State University reviews “Agent-Based Computational Economics: How the Idea Originated and Where It Is Going,” by Shu-Heng Chen. The Econlit abstract of this book begins: “Aimed at graduate students researching computational economics, experimental economics, behavioral economics, and research methodology, presents an explanation of agent-based computational economics (ACE) and how it brings decentralized procedures into market analysis, including how artificial economic agents are designed.”

Book Reviews

Agent-based computational economics (ACE) is the computational modeling of economic processes (including whole economies) as openended dynamic systems of interacting agents (Tesfatsion 2016a). In this book, author Shu-Heng Chen—an important pioneering contributor to the development of ACE—undertakes a detailed exposition of the ACE modeling approach that focuses on two fundamental questions. First, what is ACE? Second, to what extent is ACE useful, or even necessary, for understanding economic processes?

Rediscovering the Economics of Keynes in an Agent-Based Computational Setting

The aim of this paper is to use agent-based computational economics to explore the economic thinking of Keynes. Taking his starting point at the macroeconomic level, Keynes argued that economic systems are characterized by fundamental uncertainty — an uncertainty that makes rule-based behavior and reliance on monetary magnitudes more optimal to the economic agent than profit — and utility optimization in the traditional sense. Unfortunately, more systematic studies of the properties of such a system were not possible at the time of Keynes. However, the system envisioned by Keynes holds a lot of properties in common with what we today call complex dynamic systems, and we may apply the method of agent-based computational economics to his ideas. The presented agent-based Keynesian model demonstrates, as argued by Keynes, that the economy can self-organize without relying on price movement as an equilibrating factor. In our implementation, self-organization, however, does not mean a steady long run equilibrium but a tendency to generate cycles.

Evaluation of Price-sensitive Loads' Impacts on LMP and Market Power Using LMP Decomposition

This paper presents a novel approach for evaluating impacts of price-sensitive loads on electricity price and market power. To accomplish this aim an analytical method along with agent-based computational economics are used. At first, Nash equilibrium is achieved by computational approach of Q-learning then based on the optimal bidding strategies of GenCos, which are figured out by Q-learning, ISO's social welfare maximization is restated considering demand side bidding. In this research, it was demonstrated that Locational Marginal Price (LMP) at each node of system can be decomposed into five components. The first constitutive part is a constant value for the respective bus, while the next two components are related to GenCos and the last two parts are associated to Load Serving Entities (LSEs). Market regulators can acquire valuable information from the proposed LMP decomposition. First, sensitivity of electricity price at each bus and Lerner index of GenCos to the bidding strategies and maximum price- sensitive demand of LSEs are revealed through weighting coefficients of the last two terms in the decomposed LMP. Moreover, the decomposition of LMP expresses contribution of LSEs to the electricity price. The simulation results on two test systems confirm the capability of the proposed approach.

Macroeconomic Policy in DSGE and Agent-Based Models Redux: New Developments and Challenges Ahead

The Great Recession seems to be a natural experiment for economic analysis, in that it has shown the inadequacy of the predominant theoretical framework - the New Neoclassical Synthesis (NNS) - grounded on the DSGE model. In this paper, we present a critical discussion of the theoretical, empirical and political-economy pitfalls of the DSGE-based approach to policy analysis. We suggest that a more fruitful research avenue should escape the strong theoretical requirements of NNS models (e.g., equilibrium, rationality, representative agent, etc.) and consider the economy as a complex evolving system, i.e. as an ecology populated by heterogenous agents, whose far-from-equilibrium interactions continuously change the structure of the system. This is indeed the methodological core of agent-based computational economics (ACE), which is presented in this paper. We also discuss how ACE has been applied to policy analysis issues, and we provide a survey of macroeconomic policy applications (fiscal and monetary policy, bank regulation, labor market structural reforms and climate change interventions). Finally, we conclude by discussing the methodological status of ACE, as well as the problems it raises.

Evaluating the influence of increasing block tariffs in residential gas sector using agent-based computational economics

Designing a desirable increasing block tariff for the residential gas retail market has been a challenging task for regulated utilities, especially in China. To deal with such problems, in this paper, we establish an agent-based, computational economics system to provide a formal evaluation of the direct and indirect influences of several issued increasing block tariffs in the residential gas market. Moreover, a comprehensive demand response behaviour model has been improved in term of price elasticity, while still coping with income levels and complex social environment. We also compute and compare the outcomes of several increasing block tariffs with the initial flat tariff by running the system on a test-case using real-world data from a middle-scale gas retail market in Wuhan. The results indicate that there is an appropriate increasing block gas tariff scheme that has greater ability to improve social equity while still ensuring operator revenue and promoting gas conservation. In order to offset the limitations of the proposed increasing block tariffs, the regulator should adopt some complementary measures, such as applying appropriate policies targeting the intended consumers, and allowing large families to obtain extra allowance of volume.

The ACEWEM framework: An integrated agent-based and statistical modelling laboratory for repeated power auctions

An agent-based model for power auction is proposed.A new stochastic decision algorithm is designed.We study the emergence of market power.Unexpected supply or demand shocks lead to emergence of strategic offers.Re-dispatch scheme results in higher market prices compared with the LMP scheme. We propose a novel framework for experimental designs of liberalised wholesale power markets, namely the Agent-based Computational Economics of the Wholesale Electricity Market (ACEWEM) framework. Here, we describe a detailed market simulation whereby the strategies of power generators emerge as a result of a stochastic profit maximisation learning algorithm based upon the GAMLSS (Generalized Additive Models for Location Scale and Shape) statistical framework. The ACEWEM framework, which integrates the agent-based modelling paradigm with formal statistical methods to represent better real-world decision rules, is designed to be the foundation for large custom-purpose experimental studies inspired by computational learning. The paper therefore makes a methodological contribution in the development of an expert model of repeated auctions with capacity and physical constrains. It also makes an applied contribution by providing a more realistic basis for identifying whether high market prices can be ascribed to problems of market structure or exercise of market power.

Complex agent-based models: application of a constructivism in the economic research

The current state in research of economic systems is characterised by two prevailing issues. Firstly, study of economic systems is traditionally based on analytical and econometric tools, which have been the main arbiters of the veracity or plausibility of assumptions and hypotheses in economics. This approach has been proved to be highly suitable for theory development. Secondly, practical issues and necessity to support decision-making led to development of various modelling and simulation techniques or tools. However, majority of these approaches usually fail when coping with complexity. Furthermore, several main areas of interest can be identified in the business and economics modelling. Nevertheless, these areas are mostly independent due to their problem- based focusing on particular issues and their solutions. Depicted gaps might be bridged with the help of new modelling paradigms that have been established only recently. Application of agent- based modelling in the realm of economic systems is labelled as Agent-based Computational Economics (ACE). In particular sections of this paper results of experiments run on the novel model are described. The model is based on agents, which are described as a vector of several observed parameters, and four types of agents are used, namely consumer agent, factory agent, mining agent, and transportation agent. In addition, a colony is added as the fifth type of meta-agent. Scalability and configuration options of the model enable for various configuration and thus for conducting specific experiments. The presented system is already implemented as a prototype version in the NetLogo environment. The paper depicts two example scenarios, resource production and resource proximity, and offers interpretation of achieved results. Since most of the work done so far was focused on individual agents, group perspective as an important extension of ACE modelling is suggested as the further research and development direction.

Book Review: Agent-based computational economics Using NetLogo

Book Review: Agent-based computational economics Using NetLogo

Neuroeconomics and Agent-Based Computational Economics

Recently, the relation between neuroeconomics and agent-based computational economics (ACE) has become an issue concerning the agent-based economics community. Neuroeconomics can interest agent-based economists when they are inquiring for the foundation or the principle of the software-agent design. It has been shown in many studies that the design of software agents is non-trivial and can determine what will emerge from the bottom. Therefore, it has been quested for rather a period regarding whether anyone can sensibly design these software agents, including both the choice of software agent models, such as reinforcement learning, and the parameter setting associated with the chosen model, such as risk attitude. In this paper, the author will start a formal inquiry by focusing on examining the models and parameters used to build software agents.