Search

Since the first agent-based models (ABM), the scientific community has been interested in making not only the results of computational models understandable but also the modeling description, to facilitate their replication. The form that has been adopted to a greater extent has been the ODD (Overview, Design concepts, and Details) protocol, which provides a generic structure for its documentation. This protocol provides a way to clearly explain the procedures and interactions of the complex systems to be analyzed, with applications that have spread across different disciplines. This study presents a bibliometric analysis using Scopus database of the articles that emerged from the first publication of this protocol in 2006. The results show that the use of this form of documenting has grown in Economics, but its participation remains stable in relative terms. Results also show that the top journal both in number of articles and impact is JASSS, but it is not observed that there is a greater dissemination toward specific Economics journals or that it is used for a greater number of topics. Finally, it is found that there is a common core in Economics’ ABM literature, with a bias toward methodological articles in the articles that follow the ODD protocol while there is a bias towards articles on economic models in those that do not follow this protocol.

Sovereign default contagion was one of the most debated topics during the Eurozone sovereign debt crisis. Despite all the improvements in the financial situation since 2010, namely after European Central Bank quantitative easing policies, the nature of the problem and the policy prescriptions are still under dispute today. Using an agent-based model, we simulate sovereign default contagion for different monetary policy options in a world where governments have random incomes, different heterogeneous borrowing behaviors and risk aversion levels and where countries can enter into ex-ante agreements to protect themselves against default. Our simulations showed that default contagion can be a very fast and “destructive” process, and that monetary policy can have a very important role in preventing sovereign defaults through zero interest rate and quantitative easing policies.

Moral Foundation Theory states that groups of different observers may rely on partially dissimilar sets of moral foundations, thereby reaching different moral valuations. The use of functional imaging techniques has revealed a spectrum of cognitive styles with respect to the differential handling of novel or corroborating information that is correlated to political affiliation. Here we characterize the collective behavior of an agent-based model whose inter individual interactions due to information exchange in the form of opinions are in qualitative agreement with experimental neuroscience data. The main conclusion derived connects the existence of diversity in the cognitive strategies and statistics of the sets of moral foundations and suggests that this connection arises from interactions between agents. Thus a simple interacting agent model, whose interactions are in accord with empirical data on conformity and learning processes, presents statistical signatures consistent with moral judgment patterns of conservatives and liberals as obtained by survey studies of social psychology., Comment: 11 pages, 4 figures, 2 C codes, to appear in Advances in Complex Systems

This paper, describes an agent based model of the spreading of a population over a territory. The models aims at reproducing a distribution of settlements with statistical and spatial characteristics similar to a historically produced pattern. The model operates on a representation of a real territory, taking into account hydrography and relief. The two main goals are to obtain a rank size distribution of the size of settlements which corresponds to a power law (also known as the Zipf Law of settlements) and to place the settlements in the territory in patterns that are close to the real ones, in zones where settlements were the result of a long historical process. The goal of the project was to demonstrate that a set of relatively simple rules could produce a complex pattern, similar to the result of a long and complex historical process. Therefore, it is an assumed reductionist approach. Our conclusions show that a simple territorial logic, taking into account the quality of land, accessibility, population growth and migration preferences could reproduce Zipf distributions and interesting patterns of agent flow among the settlements created. However, achieving spatial patterns closer to the historical record needs an extra dimension involving field of sight. The best results were achieved by creating an artifical population which chooses to create settlements in places where a wide field of view exists of quality territory.

Actors in the Prisoner’s Dilemma agent-based model presented here decide between cooperation and defection in binary interactions determined by distance and overall gains. The paper thus tries to answer one of the fundamental questions of international politics: how does cooperative behavior perform in an environment governed by power and location? Two kinds of noise and the reward for mutual cooperation oscillating between temptation and punishment payoffs with a variable speed were added similarly like few completely new strategies inspired by foreign policy behavior of states. The initial success of generous reciprocal altruists is no surprise, but the lacking relationship between frequency of interactions and cooperativeness at the level of pairs already suggests some similarity with the system of states. Yet, the most important outcome is victory of the balance of threat strategy in all reruns with a heterogeneous pool of actors, despite the fact that this strategy was one of the least cooperative ones. At the same time, rules pre-selected by their success in the homogeneous and cooperative environment were still able to sustain intensive cooperation among themselves even within the heterogeneous pool of strategies.

Ethnic residential segregation is common in urban areas and is especially pronounced in metropolitan areas of the US. Many factors contribute to these residential patterns. The potential role of ethnic preferences has received increasing attention in recent years, but the development of theoretical models has been limited in many respects. This study seeks to extend the understanding of the possible role of ethnic preferences by investigating an agent-based model that introduces the concept of spatial potential — the desirability of particular spatial locations — as a competing preference. The distribution of spatial potential is varied from convex to concave distributions on the city landscape to investigate the resulting impact on patterns of segregation. Three types of zones are identified in the model city, namely the minority ghetto, the majority ring, and the integrated ring. The present model is considered to be an example of a discrete dynamic system that is characterized by its random move condition and its definition on the grid system. Insights from analytic approaches are explored in the context of the new dynamics to gain a better understanding of the emergent patterns of the system.

Do the Marshallian and Walrasian models have a firm foundation on micromotives, or are they just macroabstractions that we could dispense in microeconomics? Previous evidence from experimental economics with human subjects in continuous double auction markets shows that, if the supply is downward-sloping or the demand is upward-sloping, the Marshallian stability model captures the observed phenomena but the Walrasian does not. But, of course, in human subjects experiments the basic question of what the human agents' behavior is remains open. We build an artificial agent-based model to show that the path of convergence needs more intelligence than the zero-intelligence framework. Results from experimental economics with human agents reproduced with artificial agents at a low cost and with an increase in the reliability of the experiment. Not only the existence but also the computation of the stability is relevant. These results cast a shadow on the interest of both instability concepts in economic theory.

The aim of this paper is to model the process of learning within a social network and compare the levels of learning in two different situations: one where individuals know others' competencies as given data and interact on this basis; and one where individuals know nothing about others' competencies but rather build this knowledge over time, according to their past interactions. For this purpose, we build an agent-based model, and model these two scenarios of simulations. Results are partly studied using network analysis, and they show that in the second type of simulations agents are able to identify the most competent agents in the network and increase their competencies. Results also show that learning is easier when there is no prior knowledge of others' competencies. Otherwise, agents deal with a congestion effect that slows down the learning process.

Endogenous, ideas-led growth theory and the literature on agent-based modeling with neighborhood effects are crossed. In an economic overlapping generations framework, it is shown how social interactions and neighborhood effects are of vital importance in the endogenous determination of the long run number of skilled workers and therefore of the growth prospects of an economy. Neighborhood effects interact with the initial distribution of skilled agents across space and play a key role in the long run stabilization of the number of skilled individuals. Our model implies a tendency toward segregation, with a possibly positive influence on growth, if team effects operate. The long run growth rate is also shown to depend on the rate of time preference. Initial circumstances are of vital importance for long run outcomes. A poor initial education endowment will imply a long run reduced number of skilled workers and a mediocre growth rate, so there is no economic convergence tendency. On the contrary, poor societies will grow less, or will even fall into a poverty trap, and will diverge continuously from richer ones.

Agent-based models (ABMs) have been attracting the attention of researchers in the social sciences, becoming a prominent paradigm in the study of complex social systems. Although a great number of models have been proposed for studying a variety of social phenomena, no general agent design methodology is available. Moreover, it is difficult to validate the accuracy of these models. For this reason, we believe that some guidelines for ABMs design must be devised; therefore, this paper is a first attempt to analyze the levels of ABMs, identify and classify several aspects that should be considered when designing ABMs. Through our analysis, the following implications have been found: (1) there are two levels in designing ABMs: the individual level, related to the design of the agents' internal structure, and the collective level, which concerns the design of the agent society or macro-dynamics of the model; and (2) the mechanisms of these levels strongly affect the outcomes of the models.

We model an agent-based economy in which heterogeneous agents (firms and a bank) interact in the financial markets. The heterogeneity is due to the balance sheet conditions and to size. In our simulations, at the aggregate level, output displays changes in trend and volatility giving rise to complex dynamics. The average solvency and liquidity ratios peak during recessions as empirical analysis shows. At the firm level the model generates: (i) firm sizes left-skewed distributed, (ii) growth rates Laplace distributed. Furthermore, small idiosyncratic shocks can generate large aggregate fluctuations.

The inability of investors and academics to consistently predict, and understand the behavior of financial markets has forced the search for alternative analytical frameworks. Analyzing financial markets as complex systems is a framework that has demonstrated great promises, with the use of agent-based models (ABMs) and the inclusion of network science playing an important role in increasing the relevance of the framework. Using an artificial stock market created via an ABM, this paper provides a significant insight into the mechanisms that drive the returns in financial markets, including periods of elevated prices and excess volatility. The paper demonstrates that the network topology that investors form and the dividend policy of firms significantly affect the behavior of the market. However, if investors have a bias to following their neighbors then the topology becomes redundant. By successfully addressing these issues this paper helps refine and shape a variety of additional research tasks for the use of ABMs in uncovering the dynamics of financial markets.

A model framework that describes a simple production economy is presented, in which the micro-dynamics can be projected to a closed dynamics of aggregate variables. The construction is based on an agent-based model with heterogeneity both regarding production characteristics and the strategies agents use to predict future prices as a basis for choosing what to produce. The world is divided into a number of regional markets that collect the locally produced goods and via inter-market trade supply regional demands with goods. We discuss the model features that make it possible to project the agent-based micro-dynamics to a closed form dynamics on the level of regionally aggregate quantities. One advantage of such a projection from a high dimensional agent-based dynamics to a low dimensional dynamical system is that the stability characteristics can be analytically approached, and this is illustrated by a derived condition for when a mix of naive and rational agents can stabilize the system. Some illustrations of the general framework in a model of global agricultural land-use are also given.

Simulation with agent-based models is increasingly used in the study of complex socio-technical systems and in social simulation in general. This paradigm offers a number of attractive features, namely the possibility of modeling emergent phenomena within large populations. As a consequence, often the quantity in need of calibration may be a distribution over the population whose relation with the parameters of the model is analytically intractable. Nevertheless, we can simulate. In this paper we present a simulation-based framework for the calibration of agent-based models with distributional output based on indirect inference. We illustrate our method step by step on a model of norm emergence in an online community of peer production, using data from three large Wikipedia communities. Model fit and diagnostics are discussed., 31 pages, 7 figures, accepted on Adv. Compl. Sys

We propose a simple microscopic model of consumer behavior in which decisions are made solely on the basis of person to person communication. Within our model we determine which factors are important in word of mouth marketing. We provide both Monte Carlo simulations, as well as mean-field analytical calculations and show that both the average decision time and uncertainty of decision grow nonlinearly with the amount of information up to a certain threshold value.

In this paper, an agent-based model for opinion dynamics on an adaptive coupled random network is proposed. Based on Festinger's idea of "cognitive dissonance", in the proposed model an agent can either make opinion exchange with a neighbor according to the bounded confidence mechanism, or migrate toward another network position in case that the majority of the adjacent agents are beyond the confidence bound. Through numerical simulations, we test how the key factors, such as the interconnectivity of the two communities, the confidence bound or the communal tolerance to diversity, the initial distributions of the opinions, and the level of sense of community, affect the final opinion state of the system. The overall analyses show a general picture of the dynamics of opinions on an adaptive network with community structure. In particular, the results reveal that the clustering of similar agents has a bifurcating function for the opinion dynamics. Given that the inter-communal influence is high, the clustering fosters the global consensus. If the inter-communal influence is weak, the clustering would instead intensify polarization and thus hinder the formation of global consensus. The factors of the communal tolerances and interconnectivity leverage the bifurcating effect.

In this paper, we modify a two-dimensional variant of a two-state nonlinear voter model and apply it to understand how new ideas, products or behaviors spread throughout the society in time. In particular, we want to find answers to two important questions in the field of diffusion of innovation: Why does the diffusion of innovation take sometimes so long? and Why does it fail so often? Because these kind of questions cannot be answered within classical aggregate diffusion models, like the Bass model, we use an agent-based modeling approach.

Computer simulations of complex food-webs are important tools for deepening our understanding of these systems. Yet most computer models assume, rather than generate, key system-level patterns, or use mathematical modeling approaches that make it difficult to fully account for nonlinear dynamics. In this paper, we present a computer simulation model that addresses these concerns by focusing on assumptions of agent attributes rather than agent outcomes. Our model utilizes the techniques of complex adaptive systems and agent-based modeling so that system level patterns of a marine ecosystem emerge from the interactions of thousands of individual computer agents. This methodology is validated by using this general simulation model to replicate fundamental properties of a marine ecosystem, including: (i) the predator–prey oscillations found in Lotka–Volterra; (ii) the stepped pattern of biomass accrual from resource enrichment; (iii) the Paradox of Enrichment; and (iv) Gause's Law.

In this paper, we present a new agent-based model for the simulation of tax compliance and tax evasion behavior (SIMULFIS). The main novelties of the model are the introduction of a "behavioral filter approach" to model tax decisions, the combination of a set of different mechanisms to produce tax compliance (namely rational choice, normative commitments and social influence), and the use of the concept of "fraud opportunity use rate" (FOUR) as the main behavioral outcome. After describing the model in detail, we display the main behavioral and economic results of 1,920 simulations calibrated for the Spanish case and designed to test for the internal validity of SIMULFIS. The behavioral outcomes show that scenarios with strict rational agents strongly overestimate tax evasion, while the introduction of social influence and normative commitments allows to generate more plausible compliance levels under certain deterrence conditions. Interestingly, the relative effect of social influence is shown to be ambivalent: it optimizes compliance under low and middle deterrence conditions, but not when deterrence is made harder. Finally, SIMULFIS economic outcomes are broadly in line with theoretical expectations, thus supporting the reliability of the model.

In this paper, we report on the theoretical foundations, empirical context and technical implementation of an agent-based modeling (ABM) framework, that uses a high-performance computing (HPC) approach to investigate human population dynamics on a global scale, and on evolutionary time scales. The ABM-HPC framework provides an in silico testbed to explore how short-term/small-scale patterns of individual human behavior and long-term/large-scale patterns of environmental change act together to influence human dispersal, survival and extinction scenarios. These topics are currently at the center of the Neanderthal debate, i.e., the question why Neanderthals died out during the Late Pleistocene, while modern humans dispersed over the entire globe. To tackle this and similar questions, simulations typically adopt one of two opposing approaches, top-down (equation-based) and bottom-up (agent-based) models of population dynamics. We propose HPC technology as an essential computational tool to bridge the gap between these approaches. Using the numerical simulation of worldwide human dispersals as an example, we show that integrating different levels of model hierarchy into an ABM-HPC simulation framework provides new insights into emergent properties of the model, and into the potential and limitations of agent-based versus continuum models.