Publications (111)186.87 Total impact

Article: Dynamic Stock Correlation Network
[Show abstract] [Hide abstract]
ABSTRACT: Financial markets are the outcome of highly complex interactions among a number of agents. Such a complex system possibly contains all sorts of features, e.g., not only static but also dynamic. Here we study dynamic correlations hidden in the S&P 500 market by adopting a combined method of the Complex Principal Component Analysis (CPCA) and the Random Matrix Theory (RMT). The CPCA is entirely dependent on complexification of time series using the Hilbert transformation and enables us to extract correlations between stock prices moving with different phases to one another. The RMT serves as a null hypothesis for distinguishing true correlations from noisy financial data. The extracted information on dynamic correlations of the market is projected onto a correlation network in which pairs of stocks with phase difference smaller than certain threshold are linked with strength of their correlations as weight. We then detect communities of comoving stocks in the network and also elucidate leadlag relationship between those communities.  [Show abstract] [Hide abstract]
ABSTRACT: We carry out multivariate time series analysis on price indices of individual goods and services collected over the last 35 years in Japan. Adoption of the complex principal component analysis (CPCA) enables us to have a new insight into dynamic correlation structure involved in the price data. The CPCA is based on complexification of real data using the Hilbert transformation; leadlag relations between individual prices manifest in a form of instantaneous phases of the complex time series. The correlation matrix in the CPCA is purified by adopting the random matrix theory as a null hypothesis for removal of statistical noises. We identify four significant eigenmodes for price movement which are free from seasonal variations. Each of them has different characteristics of dynamical correlations and is shown to be responsive to different economic events.  [Show abstract] [Hide abstract]
ABSTRACT: We detect the backbone of the weighted bipartite network of the Japanese credit market relationships. The backbone is detected by adapting a general method used in the investigation of weighted networks. With this approach we detect a backbone that is statistically validated against a null hypothesis of uniform diversification of loans for banks and firms. Our investigation is done year by year and it covers more than thirty years during the period from 1980 to 2011. We relate some of our findings with economic events that have characterized the Japanese credit market during the last years. The study of the time evolution of the backbone allows us to detect changes occurred in network size, fraction of credit explained, and attributes characterizing the banks and the firms present in the backbone.  [Show abstract] [Hide abstract]
ABSTRACT: We present an analysis of the credit market of Japan. The analysis is performed by investigating the bipartite network of banks and firms which is obtained by setting a link between a bank and a firm when a credit relationship is present in a given time window. In our investigation we focus on a community detection algorithm which is identifying communities composed by both banks and firms. We show that the clusters obtained by directly working on the bipartite network carry information about the networked nature of the Japanese credit market. Our analysis is performed for each calendar year during the time period from 1980 to 2011. Specifically, we obtain communities of banks and networks for each of the 32 investigated years, and we introduce a method to track the time evolution of these communities on a statistical basis. We then characterize communities by detecting the simultaneous overexpression of attributes of firms and banks. Specifically, we consider as attributes the economic sector and the geographical location of firms and the type of banks. In our 32 year long analysis we detect a persistence of the overexpression of attributes of clusters of banks and firms together with a slow dynamics of changes from some specific attributes to new ones. Our empirical observations show that the credit market in Japan is a networked market where the type of banks, geographical location of firms and banks and economic sector of the firm play a role in shaping the credit relationships between banks and firms.  [Show abstract] [Hide abstract]
ABSTRACT: We have analyzed the Indices of Industrial Production (Seasonal Adjustment Index) for a long period of 240 months (January 1988 to December 2007) to develop a deeper understanding of the economic shocks. The angular frequencies estimated using the Hilbert transformation, are almost identical for the 16 industrial sectors. Moreover, the partial phase locking was observed for the 16 sectors. These are the direct evidence of the synchronization in the Japanese business cycle. We also showed that the information of the economic shock is carried by the phase timeseries. The common shock and individual shocks are separated using phase timeseries. The former dominates the economic shock in all of 1992, 1998 and 2001. The obtained results suggest that the business cycle may be described as a dynamics of the coupled limitcycle oscillators exposed to the common shocks and random individual shocks.  [Show abstract] [Hide abstract]
ABSTRACT: Possibility of risk contagion over an entire financial system may be elucidated by studying relationship among creditors and borrowers. Here we study temporal change of community structure of a credit network formed by banks and listed firms through their financial relations over the last 30 years in Japan; the communities may be regarded as reflecting concentration of risk. The network is a bipartite graph consisting of two species of nodes connected with bidirectional links. The direction of links is identified with that of risk flows and their weights are relative loans with respect to the targets. In a partial credit network obtained only with the links pointing from firms toward banks, the city banks forms one major community in most of the time period to share risk when firms go wrong. On the other hand, a partial network only with the links from banks toward firms is decomposed into communities of similar size each of which has its own city bank. Finally we take overlapping parts of the two community sets to find cores of the risk concentration in the credit network.  [Show abstract] [Hide abstract]
ABSTRACT: We elucidate correlations among stock price movements in S&P 500 and Tokyo Stock Exchange (TSE) taking advantage of the concept of community in networks. The correlation matrix, purified by random matrix theory, is regarded as the adjacency matrix for a stock correlation network. The network thus constructed has links with weights of either sign depending on whether stocks are correlated (positive) or anticorrelated (negative). Community is defined here as a group of stocks related to each other with positive correlation coefficients. The community detection allows us to find that the stocks in S&P 500 are split up into four communities with two conflicting triangular relations. In TSE, there exists three communities of stocks forming a conflicting triangle. We thus see that the frustrated correlation structure is common to the welldeveloped financial markets.  [Show abstract] [Hide abstract]
ABSTRACT: Integration of principal component analysis (PCA) with random matrix theory (RMT) has been successful in analyzing cross correlations between stock price movements in financial markets. RMT is used as a null hypothesis to distinguish between genuine cross correlations and noises. In this paper, we develop a RMTaided complex PCA method based on the Hilbert transformation of time series. The complex data thus generated carry dynamic information in a form of instantaneous phase; the conventional PCA is entirely dependent on simultaneous correlations in time. Accordingly RMT is generalized to be adaptable to complex PCA. The data set analyzed here is daily returns in Tokyo Stock Exchange (TSE) spanning from 1996 to 2006. Diagonalization of the complex correlation matrix enables us to find that a small number of the eigenvalues certainly deviate from the RMT prediction. The largest eigenvalue represents a market mode in which all of the stock prices move in a collective way. The eigenvectors of the other remaining large eigenvalues clearly show formation of stock groups as characterized by business sectors and also indicates existence of dynamical correlations between some sectors.  [Show abstract] [Hide abstract]
ABSTRACT: We construct a theoretical model for equilibrium distribution of workers across sectors with different labor productivity, assuming that a sector can accommodate a limited number of workers which depends only on its productivity. A general formula for such distribution of productivity is obtained, using the detailbalance condition necessary for equilibrium in the EhrenfestBrillouin model. We also carry out an empirical analysis on the average number of workers in given productivity sectors on the basis of an exhaustive dataset in Japan. The theoretical formula succeeds in explaining the two distinctive observational facts in a unified way, that is, a Boltzmann distribution with negative temperature on lowtomedium productivity side and a decreasing part in a powerlaw form on high productivity side.  [Show abstract] [Hide abstract]
ABSTRACT: The electronic band calculations of noble metal halides are studied to make the high ionic conducting origin of silver and cupper ions clear using the tightbinding method. The d bands of Ag ions are much more weakly coupled with the p bands of halogen ions, while those of Cu ions are much more strongly coupled with the p bands. The strength of pd hybridization is discussed to connect with the activation energy for the ionic conduction. It is shown that the high ionic conductivity of AgX primary stems from combination of the deformability of the d shell and the weakness of the pd hybridization.  [Show abstract] [Hide abstract]
ABSTRACT: Exhaustive financial data of firms in Japan enables us to shed light on how the labor productivity, defined here as value added produced by one worker in a year, is diverse across firms and workers. Statistical equilibrium theory reinforced with the concept of negative temperature turns out to be useful to explain the empirical facts on a major part of the distribution of workers over labor productivity states, where particle and singleparticle energy are replaced by worker and labor productivity, respectively. The zerotemperature state in the negative temperature regime corresponds to the optimized state for the current mainstream economics, where all workers are allocated to a state of the highest productivity. Significant difference in temperature is observed between the manufacturing and nonmanufacturing sectors. The negative temperature in the nonmanufacturing sector is three times lower than that in the manufacturing sector, indicating that the former may suffer from a much wider demand gap. In contrast, the two sectors are almost in equilibrium with respect to exchange of workers.  [Show abstract] [Hide abstract]
ABSTRACT: We report the numerical calculations of the maximal eigenvalue for random correlation matrices which contain autocorrelations in data. Here the AR(1) model is adopted for such a study, we work out an empirical formula for autocorrelation correction of the maximal eigenvalue, which are accurate in a wide range of parameters. As an application of this formula, we propose a criterion to single out statistically meaningful correlations in the principal component analysis. The new criterion within the AR(1) model incorporates autocorrelation effects into the current method based on the random matrix theory.  [Show abstract] [Hide abstract]
ABSTRACT: We reported a correlation structure hidden in the Tokyo Stock Exchange (TSE) market at KESIDT2011. By regarding the TSE market as a network (stocks and correlation coefficients correspond to nodes and weights of links between nodes, respectively) and minimizing the Frustration among nodes, the stocks were decomposed into four comoving groups forming communities. Three of them are strongly anticorrelated to each other, and the remainder is comparatively neutral to the rest of the communities. In this paper we further extend the previous work to detect tightlycoupled groups within the communities; "Hamiltonian" is used instead of the Frustration. The Hamiltonian has two parameters which control degree of strength for correlations to be extracted. It is found that six sectors (Electric Appliance, Banks, Electric Power & Supply, Information & Communication, Securities & Commodity Futures, and Insurance) form strong cores in the communities.  [Show abstract] [Hide abstract]
ABSTRACT: We investigate a production network constructed by about 800 thousand firms in Japan with focus on its transaction flow between communities. Communities detected by maximizing modularity often contain nodes with common properties such as characterized by regions or industry sectors. Communities may thus upstreamdownstream relationship according to their characteristics. Such directional bias of the connections between communities is evaluated through a polarization matrix of the network direction. We also devise a visualization method for directed network based on physical model. We attempt to draw a map of Japanese transaction flow in viewpoint of community structure.  [Show abstract] [Hide abstract]
ABSTRACT: We approach the correlation structure in the Tokyo Stock Exchange (TSE) market through a concept of community of network. To construct a network, the correlation matrix of stock price changes, purified by random matrix theory, is regarded as an adjacency matrix. The stock correlation network thus constructed has negatively weighted links as well as positively weighted links. By extracting groups in which stocks are mainly interconnected by positive links, we find that the stocks decomposed into four comoving groups forming communities, three of which are strongly anticorrelated to each other, and the remainder is comparatively neutral to the rest of the communities. The conflicting triangle relationship between communities may cause complicated behavior in a welldevelopment market such as TSE. Additionally, it is observed that some industrial sectors form distinctive coherent groups and others are separated to competing communities. 
 [Show abstract] [Hide abstract]
ABSTRACT: It is widely recognized that teleconnections (correlations between climates at remote places) such as El Nĩno play a crucial role in understanding abnormal weather phenomena. To extract such correlations in multivariate climate data, the random matrix theory (RMT) combined with the principal component analysis (PCA) can be successfully used; the RMT has power to distinguish between statistically meaningful correlations and noises. Here we demonstrate that sea level pressure (SLP), which is one of basic meteorological measurements for teleconnections, have characteristic autocorrelations. Unfortunately the standard RMT is not able to distinguish between autocorrelations and crosscorrelations. We show that an AR(1) process contaminated with noise reproduces the autocorrelations of the SLP quite well. Then we estimate autocorrelation effects on the eigenvalue distribution of the SLP correlationmatrix, which makes the extraction procedure of genuine crosscorrelations more reliable.  [Show abstract] [Hide abstract]
ABSTRACT: We investigate a Japanese transaction network consisting ofabout 800 thousand firms (nodes) and four million business relations (links) with focus on its modular structure. Communities detected by maximizing modularity often are dominated by firms with common features or behaviors in the network, such as characterized by regions or industry sectors. However, it is well known that the modularity optimization approach has a resolution limit problem, that is, it fails in identifying fine communities buried in large communities. To unfold such hidden structures, we apply the community detection to each of subnetworks formed by isolating those communities from the whole body. Subcommunities thus identified are composed of firms with finer regions, more specified sectors or business affiliations. Also we introduce a new idea of reduced modularity matrix to measure the strength of relations between (sub)communities. 
Article: An ab initio analysis of electronic states associated with a silicon vacancy in cubic symmetry
[Show abstract] [Hide abstract]
ABSTRACT: The electronic orbitals localized in the vicinity of a vacancy in a silicon crystal are calculated by an ab initio method based on the density functional theory and analyzed in association with the elastic softening observed by the recent ultrasonic experiments, especially focused on an estimate of the electric quadrupole moments. The localized orbitals due to the existence of a vacancy show largely extended properties and the quadrupole moments calculated from the orbitals indicate the strong dependence on cell sizes up to 511 atoms in the basic cell. Asymptotic values of the quadrupole moments in the limit of large size are obtained by an extrapolating method. It is shown that the quadrupole moments are enhanced due to the extension of the orbitals and the ratio of the quadrupole moments of Γ5 and Γ3 symmetries agrees well with the value deduced from the experimental results.  [Show abstract] [Hide abstract]
ABSTRACT: The sectoral synchronization observed for the Japanese business cycle in the Indices of Industrial Production data is an example of synchronization. The stability of this synchronization under a shock, e.g., fluctuation of supply or demand, is a matter of interest in physics and economics. We consider an economic system made up of industry sectors and goods markets in order to analyze the sectoral synchronization observed for the Japanese business cycle. A coupled oscillator model that exhibits synchronization is developed based on the Kuramoto model with inertia by adding goods markets, and analytic solutions of the stationary state and the coupling strength are obtained. We simulate the effects on synchronization of a sectoral shock for systems with different price elasticities and the coupling strengths. Synchronization is reproduced as an equilibrium solution in a nearest neighbor graph. Analysis of the order parameters shows that the synchronization is stable for a finite elasticity, whereas the synchronization is broken and the oscillators behave like a giant oscillator with a certain frequency additional to the common frequency for zero elasticity.
Publication Stats
2k  Citations  
186.87  Total Impact Points  
Top Journals
Institutions

19982015

Niigata University
 • Department of Mathematics
 • Department of Physics
Niahiniigata, Niigata, Japan


19812013

The University of Tokyo
 Department of Physics
Tōkyō, Japan


2011

Yamaguchi University
Yamaguti, Yamaguchi, Japan


19881998

Argonne National Laboratory
 Division of Materials Science
Lemont, Illinois, United States


1990

University of Texas at Austin
 Institute for Fusion Studies
Austin, Texas, United States
