Advances in Complex Systems

Correlations of stocks in time have been widely studied. Both the random matrix theory approach and the graphical visualization of so-called minimum spanning trees show the clustering of stocks according to industrial sectors. Studying the correlation between stocks traded in markets of different countries, we show that the random matrix theory approach is able to separate stocks according to their geographical location, provided that they are not strongly correlated. These results are compared with the results from random time series created using the market model, where the main factor is the mean of returns of the stocks of each sector.

In this paper, we consider daily financial data from various sources (stock market indices, foreign exchange rates and bonds) and analyze their multiscaling properties by estimating the parameters of a Markov-switching multifractal (MSM) model with Lognormal volatility components. In order to see how well estimated models capture the temporal dependency of the empirical data, we estimate and compare (generalized) Hurst exponents for both empirical data and simulated MSM models. In general, the Lognormal MSM models generate "apparent" long memory in good agreement with empirical scaling provided that one uses sufficiently many volatility components. In comparison with a Binomial MSM specification [11], results are almost identical. This suggests that a parsimonious discrete specification is flexible enough and the gain from adopting the continuous Lognormal distribution is very limited.

Over the past four decades, the high-performing Asian economies (HPAE) have followed a development strategy based on the exposure of their local markets to the presence of foreign competition and on outward-oriented production. In contrast, Latin American (LATAM) economies began taking steps in this direction only in the late 1980s and early 1990s, but before this period they were more focused on the implementation of import substitution policies. These divergent paths have led to sharply different growth performances in the two regions. Yet, standard trade openness indicators fall short of portraying the peculiarity of the Asian experience, and of explaining why other emerging markets with similar characteristics have been less successful over the last 25 years. We offer an alternative perspective on this issue by exploiting recently developed indicators based on weighted network analysis. We study the evolution of the core–periphery structure of the World Trade Network (WTN) and, more specifically, the evolution of the HPAE and LATAM countries within this network. Using random walk betweenness centrality, we show that the HPAE countries are more integrated into the WTN and many of them, which were on the periphery in the 1980s, are now in the core of the network. In contrast, the LATAM economies have at best maintained their position over the 1980–2005 period, and in some cases have fallen in the ranking of centrality.

A link in a supplier–customer network is usually a creditor–debtor relationship. If a firm goes into a state of financial insolvency or bankruptcy, then firms on its upstream are affected secondarily along the links. By using 10-year data from recent data on bankruptcy in Japan, we show that this "link effect" is by no means negligible in a nationwide economy. While the total debt of bankruptcy typically amounts to as much as a few percent of GDP in Japan, nearly 20% of all bankruptcies are due to the link effect. Interestingly, we find that such a link effect becomes comparable with other causes of bankruptcy, including poor performance in business (namely solo failure) with respect to the number of events, as the bankruptcy is larger. This means that the link effect grows for larger bankruptcies. Because the supplier–customer network has a heavy tail in its degree distribution, the ripple effect due to the bankruptcy chain is considerable. Since every firm is embedded in a giant network of production, this study would suggest the importance of understanding the heterogeneity in the large-scale network of suppliers and customers on a nationwide scale.

In this paper, starting from Jaynes' MaxEnt methodology [10, 11], we follow the original idea of Aoki [1] to implement a canonical MaxEnt inference model for the replication of industrial firms' dynamics over a space of economic states. We develop an aggregate model to infer the distributions of agents at meso level using representative states. In particular, we estimate the access probability for agents in different states consistently with macroscopic economic constraints. The model is calibrated on the basis of a sample of firms, drawn from the AMADEUS database, within the manufacturing industry made up of nine sectors of economic activity from 1995 to 2004, and results come to experimental proof at aggregate macroscopic level.

This paper discusses the debate between those advocating a computational and those advocating a dynamic definition of complexity, and how this relates to issues in econophysics. It then reviews the criticisms that have been raised about ways in which econophysics has been done, noting that many of these are now being dealt with. Finally, it argues that while an obvious way to resolve many of these matters is to have economists and physicists work together, the physicists should be sure to work with economists who understand the complexity critique of conventional economic theory and are thus not led astray into building models that have some of the problems of standard economics models that most econophysicists have striven to overcome.

In this paper, an adaptive investment strategy for environments with periodic returns on investment is presented. In this approach, an investment model is considered where the agent decides at every time step the proportion of wealth to invest in a risky asset, keeping the rest of the budget in a risk-free asset. Every investment is evaluated in the market via stylized return on investment function (RoI), which is modeled by a stochastic process with unknown periodicities and levels of noise. For comparison, two reference strategies are presented which represent the case of agents with zero knowledge and complete knowledge of the dynamics of the returns. An investment strategy based on technical analysis to forecast the next return is also considered. To account for the performance of the different strategies, some computer experiments are performed to calculate the average budget that can be obtained with them over a certain number of time steps. To assure fair comparisons, the parameters of each strategy are first tuned for budget maximization. Afterward, the performance of these strategies is compared for RoI's with different periodicities and levels of noise.

We study the emergence of self-organized traffic flow conventions in a multiagent system. A transport system is modeled as a discrete grid, on which agents governed by a neural network can move with a velocity that depends on the quantity and directions of the agents occupying the site of the grid. We show that evolution leads to traffic flow conventions that depend on the transport system geometry. We also show that if we split the agents into different "species" that move simultaneously on the transport system but that cannot exchange genetic information, their mutual influence leads to the same traffic flow convention for all the species. Finally, we develop a very simple evolutionary dynamic model (solvable both numerically and analytically) that gives an interpretation of our results.