Narrow Results

Based on the concepts of educational psychology, sociology and statistical physics, a mathematical model for a new type of social learning process that takes place when individuals interact via the Internet is proposed and studied. The noise of the interaction (misunderstandings, lack of well organized participative activities, etc.) dramatically restricts the number of individuals that can be efficiently in mutual contact and drives phase transitions between "ordered states" such as the achievements of the individuals are satisfactory and "disordered states" with negligible achievements.

Language emergence and evolution have recently gained growing attention through multi-agent models and mathematical frameworks to study their behavior. Here we investigate further the Naming Game, a model able to account for the emergence of a shared vocabulary of form-meaning associations through social/cultural learning. Due to the simplicity of both the structure of the agents and their interaction rules, the dynamics of this model can be analyzed in great detail using numerical simulations and analytical arguments. This paper first reviews some existing results and then presents a new overall understanding.

A new stochastic stock price model of stock markets based on the contact process of the statistical physics systems is presented in this paper, where the contact model is a continuous time Markov process, one interpretation of this model is as a model for the spread of an infection. Through this model, the statistical properties of Shanghai Stock Exchange (SSE) and Shenzhen Stock Exchange (SZSE) are studied. In the present paper, the data of SSE Composite Index and the data of SZSE Component Index are analyzed, and the corresponding simulation is made by the computer computation. Further, we investigate the statistical properties, fat-tail phenomena, the power-law distributions, and the long memory of returns for these indices. The techniques of skewness–kurtosis test, Kolmogorov–Smirnov test, and R/S analysis are applied to study the fluctuation characters of the stock price returns.

Web encounter facilitate contacts between people from different communities outside space and time. Implicit Community Structure is exhibited because of highly connected links within community and sparse encounters between communities. Considering the imperceptible influence of encounter on opinions, Sznajd updating rules are used to mimic people's behaviors after encountering a stranger in another community. We introduce a model for opinion evolution, in which the interconnectivity between different communities is represented as encounter frequency, and leadership is introduced to control the strength of community's opinion guide. In this scenario, the effects of Implicit Community Structure of contact network on opinion evolution, for asymmetric and random initial distribution but with heterogeneous opinion guide, are investigated respectively. It is shown that large encounter frequency favors consensus of the whole populations and successful opinion spreading, which is qualitatively agree with the results observed in Majority model defined on substrates with predefined community structure.

Community structure is another important feature besides small-world and scale-free property of complex networks. Communities can be coupled through specific fixed links between nodes, or occasional encounter behavior. We introduce a model for opinion evolution with multiple cluster-coupled patterns, in which the interconnectivity denotes the coupled degree of communities by fixed links, and encounter frequency controls the coupled degree of communities by encounter behaviors. Considering the complicated cognitive system of people, the CODA (continuous opinions and discrete actions) update rules are used to mimic how people update their decisions after interacting with someone. It is shown that, large interconnectivity and encounter frequency both can promote consensus, reduce competition between communities and propagate some opinion successfully across the whole population. Encounter frequency is better than interconnectivity at facilitating the consensus of decisions. When the degree of social cohesion is same, small interconnectivity has better effects on lessening the competence between communities than small encounter frequency does, while large encounter frequency can make the greater degree of agreement across the whole populations than large interconnectivity can.

In this paper, we extend some ideas of statistical physics to describe the properties of human mobility. From a physical point of view, we consider the statistical empirical laws of private cars mobility, taking advantage of a GPS database which contains a sampling of the individual trajectories of 2% of the whole vehicle population in an Italian region. Our aim is to discover possible "universal laws" that can be related to the dynamical cognitive features of individuals. Analyzing the empirical trip length distribution we study if the travel time can be used as universal cost function in a mesoscopic model of mobility. We discuss the implications of the elapsed times distribution between successive trips that shows an underlying Benford's law, and we study the rank distribution of the average visitation frequency to understand how people organize their daily agenda. We also propose simple stochastic models to suggest possible explanations of the empirical observations and we compare our results with analogous results on statistical properties of human mobility presented in the literature.

The sampling method has been paid much attention in the field of complex network in general and statistical physics in particular. This paper proposes two new sampling methods based on the idea that a small part of vertices with high node degree could possess the most structure information of a complex network. The two proposed sampling methods are efficient in sampling high degree nodes so that they would be useful even if the sampling rate is low, which means cost-efficient. The first new sampling method is developed on the basis of the widely used stratified random sampling (SRS) method and the second one improves the famous snowball sampling (SBS) method. In order to demonstrate the validity and accuracy of two new sampling methods, we compare them with the existing sampling methods in three commonly used simulation networks that are scale-free network, random network, small-world network, and also in two real networks. The experimental results illustrate that the two proposed sampling methods perform much better than the existing sampling methods in terms of achieving the true network structure characteristics reflected by clustering coefficient, Bonacich centrality and average path length, especially when the sampling rate is low.

Financial market is a complex evolved dynamic system with high volatilities and noises, and the modeling and analyzing of financial time series are regarded as the rather challenging tasks in financial research. In this work, by applying the Potts dynamic system, a random agent-based financial time series model is developed in an attempt to uncover the empirical laws in finance, where the Potts model is introduced to imitate the trading interactions among the investing agents. Based on the computer simulation in conjunction with the statistical analysis and the nonlinear analysis, we present numerical research to investigate the fluctuation behaviors of the proposed time series model. Furthermore, in order to get a robust conclusion, we consider the daily returns of Shanghai Composite Index and Shenzhen Component Index, and the comparison analysis of return behaviors between the simulation data and the actual data is exhibited.

A financial time series model is developed and investigated by the oriented percolation system (one of the statistical physics systems). The nonlinear and statistical behaviors of the return interval time series are studied for the proposed model and the real stock market by applying visibility graph (VG) and multifractal detrended fluctuation analysis (MF-DFA). We investigate the fluctuation behaviors of return intervals of the model for different parameter settings, and also comparatively study these fluctuation patterns with those of the real financial data for different threshold values. The empirical research of this work exhibits the multifractal features for the corresponding financial time series. Further, the VGs deviated from both of the simulated data and the real data show the behaviors of small-world, hierarchy, high clustering and power-law tail for the degree distributions.

For a homogeneous system divisible into identical, weakly interacting subsystems, the multicanonical procedure can be accelerated if it is first applied to determine the density of states for a single subsystem. This result is then employed to approximate the state density of a subsystem with twice the size that forms the starting point of a new multicanonical iteration. Since this compound subsystem interacts less on average with its environment, iterating this sequence of steps rapidly generates the state density of the full system.

A sampling procedure for the transition matrix Monte Carlo method is introduced that generates the density of states function over a wide parameter range with minimal coding effort.

In this paper, we consider a model of relativistic networks, a topological extension of the model of relativistic particles. Numerical experiments are performed to study thermodynamical properties of the model and their relationship with explicit symmetry of solutions under time reversal. An efficient algorithm is constructed, allowing to generate numerical solutions of high complexity in the given model. The algorithm includes a generator of random topology, an optimal choice of stiffness coefficients for the network and a solver for constrained optimization problem, describing an equilibrium of the network. A system, studied in the given paper, contains about 100 thousands of equations and inequalities. Possible extensions of the algorithm are discussed, necessary for processing of relativistic networks of higher complexity, containing millions of equations.

A random stock price model based on the multi-continuum percolation system is developed to investigate the nonlinear dynamics of stock price volatility duration, in an attempt to explain various statistical facts found in financial data, and have a deeper understanding of mechanisms in the financial market. The continuum percolation system is usually referred to be a random coverage process or a Boolean model, it is a member of a class of statistical physics systems. In this paper, the multi-continuum percolation (with different values of radius) is employed to model and reproduce the dispersal of information among the investors. To testify the rationality of the proposed model, the nonlinear analyses of return volatility duration series are preformed by multifractal detrending moving average analysis and Zipf analysis. The comparison empirical results indicate the similar nonlinear behaviors for the proposed model and the actual Chinese stock market.

We demonstrate that a series of procedures for increasing the efficiency of transition matrix calculations can be realized by integrating the standard single-spin flip transition matrix method with global cluster flipping techniques. Our calculations employ a simple and accurate method based on detailed balance for computing the density of states from the Ising model transition matrix.

We demonstrate that a temperature schedule for single-spin flip transition matrix calculations can be simply and rapidly generated by monitoring the average size of the Wolff clusters at a set of discrete temperatures. Optimizing this schedule yields a potentially interesting quantity related to the fractal structure of Ising clusters. We also introduce a technique in which the transition matrix is constructed at a sequence of discrete temperatures at which Wolff cluster reversals are alternated with certain series of single-spin flip steps. The single spin-flip transitions are then employed to construct a single transition matrix.

Financial markets have been known to exhibit plentiful nonlinear complex volatility behaviors. In order to reproduce the volatility dynamics of financial price changes, the agent-based financial model is established by stochastic finite-range exclusion process. The exclusion process is a kind of statistical physics system, which is considered as modeling particle Markov motion with conserved number of particles. To measure the volatility of financial return series, a novel statistic called maximum monotonic volatility rate is put forward to measure the speed of monotonic volatility of returns. Meanwhile, average monotonic volatility duration of returns is also investigated, which can reflect the average volatility level. For verifying the rationality of the model, matching energy analysis that can detect chaos and complexity in nonlinear time series is applied to study the new statistics. Further, empirical mode decomposition and multifractal are employed to study the behaviors of monotonic volatility duration. The model has similar complexity behaviors with real markets in terms of monotonic volatility with matching energy analysis, and the proposed financial model and real markets both show multifractal and anti-correlation for average monotonic volatility series by MFDFA method. The results display that the model is feasible with respect to above volatility analyses.

As an elementary task in statistical physics and network science, link prediction has attracted great attention of researchers from many fields. While numerous similarity-based indices have been designed for undirected networks, link prediction in directed networks has not been thoroughly studied yet. Among several representative works, motif predictors such as “feed-forward-loop” and Bi-fan predictor perform well in both accuracy and efficiency. Nevertheless, they fail to explicitly explain the linkage motivation of nodes, nor do they consider the unequal contributions of different neighbors between node pairs. In this paper, motivated by the investment theory in economics, we propose a universal and explicable model to quantify the contributions of nodes on driving link formation. Based on the analysis on two typical investment relationships, namely “follow-up” and “co-follow”, an investment-profit index is designed for link prediction in directed networks. Empirical studies on 12 static networks and four temporal networks show that the proposed method outperforms eight mainstream baselines under three standard metrics. As a quasi-local index, it is also suitable for large-scale networks.

Text as a complex system is commonly studied by various methods, like complex networks or time series analysis, in order to discover its properties. One of the most important properties of each text is its keywords, which are extracted by word ranking methods. There are various methods to rank words of a text. Each method differently ranks words according to their frequency, spatial distribution or other word properties. Here, we aimed to explore how similar various word ranking methods are. For this purpose, we studied the rank correlation of some important word ranking methods for number of sample texts with different subjects and text sizes. We found that by increasing text size the correlation between ranking methods grows. It means that as the text size increases, the associated word ranks calculated by different ranking methods converge. Also, we found out that the rank correlations of word ranking methods approach their maximum value in the case of large enough texts.

From the official report of narco-war-related casualties in Mexico from December 2006 to September 2011, we show that the inter-event time distribution (calculated for a range of minimum sizes) approximately obeys simple scaling laws similar to violent conflicts in Iraq (2003–2005), Afghanistan (2008–2010) and Northern Ireland (1969–2001). Furthermore, normalizing deaths by population municipalities (the smallest Mexican political entities) yields even better fitting results.

We calculate the quantum statistical force acting on a partition wall that divides a one-dimensional box into two halves. The two half boxes containing the same (fixed) number of noninteracting bosons are kept at the same temperature, and admit the same boundary conditions at the outer walls; the only difference is the distinct boundary conditions imposed at the two sides of the partition wall. The net force acting on the partition wall is nonzero at zero temperature and remains almost constant for low temperatures. As the temperature increases, the force starts to decrease considerably, but after reaching a minimum it starts to increase, and tends to infinity with a square-root-of-temperature asympotics. This example demonstrates clearly that distinct boundary conditions cause remarkable physical effects for quantum systems.