Professor Thomas Lux, University of Kiel
Estimation of agent-based models using sequential Monte Carlo methods
Abstract: Estimation of agent-based models is currently an intense area of research. Recent contributions have to a large extent resorted to simulation-based methods mostly using some form of simulated method of moments estimation (SMM). There is, however, an entire branch of statistical methods that should appear promising, but has to our knowledge never been applied so far to estimate agent-based models in economics and finance: Markov chain Monte Carlo methods designed for state space models or models with latent variables. This latter class of models seems particularly relevant as agent-based models typically consist of some latent and some observable variables since not all the characteristics of agents would mostly be observable. Indeed, one might often not only be interested in estimating the parameters of a model, but also to infer the time development of some latent variable. However, agent-based models when interpreted as latent variable models would be typically characterized by non-linear dynamics and non-Gaussian fluctuations and, thus, would require a computational approach to statistical inference. Here we resort to Sequential Monte Carlo (SMC) estimation based on a particle filter. This approach is used here to numerically approximate the conditional densities that enter into the likelihood function of the problem. With this approximation we simultaneously obtain parameter estimates and filtered state probabilities for the unobservable variable(s) that drive(s) the dynamics of the observable time series. In our examples, the observable series will be asset returns (or prices) while the unobservable variables will be some measure of agents' aggregate sentiment. We apply SMC to two selected agent-based models of speculative dynamics with somewhat different flavor. The empirical application to a selection of financial data includes an explicit comparison of the goodness-of-fit of both models.
Professor Simone Alfarano, Universitat Jaume I
Analytical approaches to agent-based models
Abstract: Agent-based modelling is nowadays an established methodology to address the main problems in several areas in Economics: from the financial stability of the banking system, to the sustainability of the environment and its interaction with the economic resources. More and more complex computational platforms are proposed as a kind of virtual realities to test and simulate the impact of various possible economic scenarios. The common feature of those computational models is the presence of a large number of heterogeneous and interacting constituent units. This means that an extensive parameter N must be an important characteristic of those models, indicating the number of sub-units or agents. Typically, such computational models generate aggregate endogenous fluctuations, as emergent properties of the aggregation of idiosyncratic shocks affecting each individual component. The underlying mechanism responsible of such emergent properties is often unknown or just partially understood. In this talk, I illustrate two examples of analytically solvable agent-based models, where the connection between the individual shocks, the interactions among the constituent units and the aggregate fluctuations is understood in all its aspects. In particular, I will show (i) an agent based model able to describe financial time series returns and option pricing and (ii) a macro economic model which can reproduce the properties of business cycle fluctuations. In both models I will stress the importance of the parameter N in the determination of the aggregate fluctuations.
Invited Special Session
Validation of Agent-Based Models
Organizer: Professor Thomas Lux, University of Kiel
Outline: The special session will cover papers on statistical inference and other forms of validation of agent-based models. Both papers with a methodologicaL focus and papers of more applied nature will be presented.
Jiri Kukacka and Ladislav Kristoufek: Multifractality Properties of the Agent-Based Models of Financial Markets
Rama Cont and Amir Sani: Transfer Calibration for Agent-Based Models
Philipp Mundt and Yoshiyuki Arata: The formation of customer-supplier relationships: evidence from Japan
Saskia Ter Ellen, Cars Hommes and Remco Zwinkels: Comparing Behavioral Heterogeneity Across Asset Classes
Tae-Seok Jang and Stephen Sacht: Forecast Heuristics, Consumer Expectations, and New Keynesian Macroeconomics: A Horse Race
Cryptocurrency Special Session
A Cryptocurrency Fraud Spill in Japan
Prof. Naoyuki Iwashita, Kyoto University
Abstract: At midnight on January 26 of 2018, crypto-currency NEM equivalent of 58 billion yen was illegally remitted and leaked to someone. One of the major crypto-currency exchange companies "Coincheck" had stored this crypto-currency owned by customers. Coincheck managed the all of stored NEM with only one secret key. The attacker used this key illegally and issued commands to rewrite the ownership information to remit NEM to the attacker's address. Eventually, the leaked NEM has laundered anonymously into other crypto-currencies and lost from the investigation. Japan has enforced the law regulating crypto-currency ahead of other countries and has operated the registration system of exchange companies. However, the current Crypto-currency Act in Japan does not have a sufficient mechanism for customer protection that considers exchange companies store a large amount of customers' asset. The government should strengthen the legal regulation from the perspective of consumer protection. Each exchange company should consider institutional security measures using trust law and insurance. The industry needs to make an effort to set self-regulation on unified security standards and disclosure of security management and corporate governance.
May Crypto-Currency Impact Real Economy?
Dr. Hiwon Yoon, CMD Laboratory, Tokyo, Japan
Abstract: We used to use money without any suspicion normally. On the other hand crypto-currency has been emerging all over the world recently, and some people may use crypto-currency with suspicion. Under this circumstance, we need to evaluate whether it works or not from the technology point of view, and moreover we should re-recognize what money is. I will review money history, and then suggest possible future of money in the new history.
Page in preparation (incomplete). Will be updated at a later time.