EZ模型中的有限尺寸效应

研究EZ模型中的有限尺寸效应.当经纪人数目N足够大及发生交易的概率a1/N，发现有限尺 寸效应是重要的.此时，系统几乎变成包含所有经纪人的单一集团.而对较小集团，尺寸分布 仍然服从幂函数律，但是指数因涨落效应而改变.但当a1/N时，可以论证涨落效应不重要 ，因而平均场理论是严格成立的. 关键词： EZ模型 有限尺寸效应 涨落 平均场理论     

Glassy effects in the swelling/collapse dynamics of homogeneous polymers

We investigate, using numerical simulations and analytical arguments, a simple one-dimensional model for the swelling or the collapse of a closed polymer chain of size N, representing the dynamical evolution of a polymer in a Θ-solvent that is rapidly changed into a good solvent (swelling) or a bad solvent (collapse). In the case of swelling, the density profile for intermediate times is parabolic and expands in space as t ^1/3, as predicted by a Flory-like continuum theory. The dynamics slows down after a time ∝N ² when the chain becomes stretched, and the polymer gets stuck in metastable “zig-zag” configurations, from which it escapes through thermal activation. The size of the polymer in the final stages is found to grow as . In the case of collapse, the chain very quickly (after a time of order unity) breaks up into clusters of monomers (“pearls”). The evolution of the chain then proceeds through a slow growth of the size of these metastable clusters, again evolving as the logarithm of time. We enumerate the total number of metastable states as a function of the extension of the chain, and deduce from this computation that the radius of the chain should decrease as 1/ln(ln t). We compute the total number of metastable states with a given value of the energy, and find that the complexity is non-zero for arbitrary low energies. We also obtain the distribution of cluster sizes, that we compare to simple “cut-in-two” coalescence models. Finally, we determine the aging properties of the dynamical structure. The subaging behaviour that we find is attributed to the tail of the distribution at small cluster sizes, corresponding to anomalously “fast” clusters (as compared to the average). We argue that this mechanism for subaging might hold in other slowly coarsening systems. Received 23 October 2000     

Complex spatial organization in a simple model of resource allocation

A dynamical model for the distribution of resources among competing agents is studied. The model is exactly solvable in the case of global competition, which leads to the accumulation of all the resources by the agent with the highest performance. On the other hand, local competition allows for a wider resource distribution, with a much weaker correlation with individual performances. Multiplicative processes give rise to almost-ordered spatial structures, through the enhancement of random fluctuations. Received 17 August 2000 and Received in final form 1st November 2000     

Ultrafast dynamics in the excited hydrogen atom transfer states of ammonia clusters

Neutral ammonia clusters (NH₃)_m are photo-excited to the electronic state by a deep UV femtosecond laser pump pulse. Within a few hundred femtoseconds a significant fraction of the clusters rearrange to form an H-transfer state (NH₃)_m-2NH₄(3s)NH₂ with the subunit NH₄ in its 3s electronic ground state. This state is then electronically excited by a time-delayed infrared control pulse of variable wavelength. Finally, a third (probe) pulse in the UV ionizes the clusters for detection. The lifetime of the excited (NH₃)_m-2NH₄(3p)NH₂ states is found to vary between 2.7 and 0.13 ps depending on cluster size and excitation energy. It increases drastically upon deuteration. The corresponding cluster size-dependent photoelectron spectra allow us to disentangle the underlying energetics of the excitation and ionization process and reveal additional processes, such as nonresonant ionization or dissociative ionization. The experimental findings suggest that the excited H-transfer ammonia complexes with m > 2 are deactivated by an internal conversion process back to the electronically lowest H-transfer state followed by fast dissociation. Received 22 September 2001 and Received in final form 31 January 2002     

Isomerization and dissociation of small (CH <Emphasis Type="Bold">3</Emphasis>CN) <Emphasis Type="Bold">n</Emphasis> molecular clusters: a computational study

The isomerization and evaporation processes in the neutral homogeneous (CH₃CN)_n molecular clusters (n = 2-7) have been investigated using classical molecular dynamics simulations. The evaporation rate constants and the kinetic energy release in the dissociation have been analysed as a function of the cluster size and as a function of the internal energy in the parent cluster. The competition between monomer and dimer ejections has been also carefully studied. All the dynamical properties in these dissociative processes have been discussed in relation to the static properties of the clusters involved in the dissociation and also in relation to the solid-liquid like transition which appears in these homogeneous molecular clusters. Received 19 November 2002 / Received in final form 5 February 2003 Published online 29 April 2003 RID="a" ID="a"e-mail: pascal.parneix@ppm.u-psud.fr RID="b" ID="b"Laboratoire associé à l'université Paris-Sud.     

Directed spiral site percolation on the square lattice

A new site percolation model, directed spiral percolation (DSP), under both directional and rotational (spiral) constraints is studied numerically on the square lattice. The critical percolation threshold p _c ≈ 0.655 is found between the directed and spiral percolation thresholds. Infinite percolation clusters are fractals of dimension d _f ≈ 1.733. The clusters generated are anisotropic. Due to the rotational constraint, the cluster growth is deviated from that expected due to the directional constraint. Connectivity lengths, one along the elongation of the cluster and the other perpendicular to it, diverge as p → p _c with different critical exponents. The clusters are less anisotropic than the directed percolation clusters. Different moments of the cluster size distribution P _s(p) show power law behaviour with | p - p _c| in the critical regime with appropriate critical exponents. The values of the critical exponents are estimated and found to be very different from those obtained in other percolation models. The proposed DSP model thus belongs to a new universality class. A scaling theory has been developed for the cluster related quantities. The critical exponents satisfy the scaling relations including the hyperscaling which is violated in directed percolation. A reasonable data collapse is observed in favour of the assumed scaling function form of P _s(p). The results obtained are in good agreement with other model calculations. Received 10 November 2002 / Received in final form 20 February 2003 Published online 23 May 2003 RID="a" ID="a"e-mail: santra@iitg.ernet.in     

Dynamical scaling and kinetic roughening of single valued fronts propagating in fractal media

We consider the dynamical scaling and kinetic roughening of single-valued interfaces propagating in 2D fractal media. Assuming that the nearest-neighbor height difference distribution function of the fronts obeys Lévy statistics with a well-defined algebraic decay exponent, we consider the generalized scaling forms and derive analytic expressions for the local scaling exponents. We show that the kinetic roughening of the interfaces displays intrinsic anomalous scaling and multiscaling in the relevant correlation functions. We test the predictions of the scaling theory with a variety of well-known models which produce fractal growth structures. Results are in excellent agreement with theory. For some models, we find interesting crossover behavior related to large-scale structural instabilities of the growing aggregates. Received 22 May 2002 Published online 19 November 2002     

Cluster size distribution of infection in a system of mobile agents

Clusters of infected individuals are defined on data from health laboratories, but this quantity has not been defined and characterized by epidemy models on statistical physics. For a system of mobile agents we simulate a model of infection without immunization and show that all the moments of the cluster size distribution at the critical rate of infection are characterized by only one exponent, which is the same exponent that determines the behavior of the total number of infected agents. No giant cluster survives independent of the magnitude of the rate of infection.     

From market games to real-world markets

This paper uses the development of multi-agent market models to present a unified approach to the joint questions of how financial market movements may be simulated, predicted, and hedged against. We first present the results of agent-based market simulations in which traders equipped with simple buy/sell strategies and limited information compete in speculatory trading. We examine the effect of different market clearing mechanisms and show that implementation of a simple Walrasian auction leads to unstable market dynamics. We then show that a more realistic out-of-equilibrium clearing process leads to dynamics that closely resemble real financial movements, with fat-tailed price increments, clustered volatility and high volume autocorrelation. We then show that replacing the `synthetic' price history used by these simulations with data taken from real financial time-series leads to the remarkable result that the agents can collectively learn to identify moments in the market where profit is attainable. Hence on real financial data, the system as a whole can perform better than random. We then employ the formalism of Bouchaud in conjunction with agent based models to show that in general risk cannot be eliminated from trading with these models. We also show that, in the presence of transaction costs, the risk of option writing is greatly increased. This risk, and the costs, can however be reduced through the use of a delta-hedging strategy with modified, time-dependent volatility structure. Received 30 August 2000     

Models for the size distribution of businesses in a price driven market

A microscopic model of aggregation and fragmentation is introduced to investigate the size distribution of businesses. In the model, businesses are constrained to comply with the market price, as expected by the customers, while customers can only buy at the prices offered by the businesses. We show numerically and analytically that the size distribution scales like a power-law. A mean-field version of our model is also introduced and we determine for which value of the parameters the mean-field model agrees with the microscopic model. We discuss to what extent our simple model and its results compare with empirical data on company sizes in the US and debt sizes in Japan. Finally, possible extensions of the mean-field model are discussed, to cope with other empirical data. Received 10 January 2001 and Received in final form 12 March 2001     

Replica-symmetry breaking in dynamical glasses

Systems of globally coupled logistic maps (GCLM) can display complex collective behaviour characterized by the formation of synchronous clusters. In the dynamical clustering regime, such systems possess a large number of coexisting attractors and might be viewed as dynamical glasses. Glass properties of GCLM in the thermodynamical limit of large system sizes N are investigated. Replicas, representing orbits that start from various initial conditions, are introduced and distributions of their overlaps are numerically determined. We show that for fixed-field ensembles of initial conditions all attractors of the system become identical in the thermodynamical limit up to variations of order 1/, and thus replica symmetry is recovered for N→∞. In contrast to this, when fluctuating-field ensembles of initial conditions are chosen, replica symmetry remains broken in the thermodynamical limit. Received 9 July 2001     

Influence of ice concentration and floe-size distribution on cluster formation in sea-ice floes

At medium ice concentrations, sea ice consists of separate floes of different sizes interacting with each other through inelastic collisions, in a way similar to two-dimensional polydisperse granular gases. The dynamics of this type of ice cover is poorly understood. In this paper, a molecular-dynamics sea-ice model based on simplified momentum equations and a hard-disk collision model is used to analyze processes of cluster formation in sea-ice floes. The clusters, formed due to size-dependent equilibrium velocities of floes under a given forcing, have statistical properties dependent on the average ice concentration and on the parameters of the floe-size distribution. In particular, in terms of the size of the largest cluster in the system, two regimes are observed: one at low and one at high ice concentration. At high ice concentration, the dominating cluster spans the entire model domain and contains the majority of floes. The exponent of the cluster-size distribution increases with increasing exponent of the floe-size distribution. The results are discussed from the point of view of the collisional contribution to the internal stress in the ice, as well as from the role of clustering in the floe-formation processes. Thus, they may contribute to the formulation of more reliable sea-ice rheology models valid at medium ice concentrations.     

Modelling conflicts with cluster dynamics in networks

We introduce cluster dynamical models of conflicts in which only the largest cluster can be involved in an action. This mimics the situations in which an attack is planned by a central body, and the largest attack force is used. We study the model in its annealed random graph version, on a fixed network, and on a network evolving through the actions. The sizes of actions are distributed with a power-law tail, however, the exponent is non-universal and depends on the frequency of actions and sparseness of the available connections between units. Allowing the network reconstruction over time in a self-organized manner, e.g., by adding the links based on previous liaisons between units, we find that the power-law exponent depends on the evolution time of the network. Its lower limit is given by the universal value 5/2, derived analytically for the case of random fragmentation processes. In the temporal patterns behind the size of actions we find long-range correlations in the time series of the number of clusters and the non-trivial distribution of time that a unit waits between two actions. In the case of an evolving network the distribution develops a power-law tail, indicating that through repeated actions, the system develops an internal structure with a hierarchy of units.     

Probabilistic properties of neuron spiking time-series obtained in vivo

Probabilistic properties of spiking time-series obtained in vivo from singular neurons belonging to Red Nucleus of brain are analyzed for two groups of rats: genetically defined rat model of depression (Flinders Sensitive Rat Line - FSL) and a control (healthy) group. The FSL group shows a distribution of interspike intervals with a much longer tail than that found for normal rats. The former distribution (for the FSL group) indicates a power-law with exponent α = - 1±0.1. A simple thermodynamic (noise) model is elaborated to explain obtained results. Received 13 May 2001 and Received in final form 20 September 2001     

Stable power laws in variable economies; Lotka-Volterra implies Pareto-Zipf

In recent years we have found that logistic systems of the Generalized Lotka-Volterra type (GLV) describing statistical systems of auto-catalytic elements posses power law distributions of the Pareto-Zipf type. In particular, when applied to economic systems, GLV leads to power laws in the relative individual wealth distribution and in market returns. These power laws and their exponent α are invariant to arbitrary variations in the total wealth of the system and to other endogenously and exogenously induced variations. Received 31 December 2001     

Lyapunov exponent, generalized entropies and fractal dimensions of hot drops

We calculate the maximal Lyapunov exponent, the generalized entropies, the asymptotic distance between nearby trajectories and the fractal dimensions for a finite two-dimensional system at different initial excitation energies. We show that these quantities have a maximum at about the same excitation energy. The presence of this maximum indicates the transition from a chaotic regime to a more regular one. In the chaotic regime the system is composed mainly of a liquid drop while the regular one corresponds to almost freely flowing particles and small clusters. At the transitional excitation energy the fractal dimensions are similar to those estimated from the Fisher model for a liquid-gas phase transition at the critical point. Received: 16 March 2001 / Accepted: 12 July 2001     

Dynamical properties of constrained drops

In this communication we analyze the behavior of excited drops contained in spherical volumes. We study different properties of the dynamical systems, i.e. the maximum Lyapunov exponent MLE, the asymptotic distance in momentum space d _∞ and the normalized variance of the maximum fragment. It is shown that the constrained system behaves as undergoing a first-order phase transition at low densities while as a second-order one at high densities. The transition from liquid-like to vapor-like behavior is signaled both by the caloric curves, the thermal response functions and the MLE. The relationship between the MLE, d _∞, and the caloric curve is explored. Received: 28 March 2002 / Accepted: 17 May 2002     

Front propagation in chaotic and noisy reaction-diffusion systems: a discrete-time map approach

We study the front propagation in reaction-diffusion systems whose reaction dynamics exhibits an unstable fixed point and chaotic or noisy behaviour. We have examined the influence of chaos and noise on the front propagation speed and on the wandering of the front around its average position. Assuming that the reaction term acts periodically in an impulsive way, the dynamical evolution of the system can be written as the convolution between a spatial propagator and a discrete-time map acting locally. This approach allows us to perform accurate numerical analysis. They reveal that in the pulled regime the front speed is basically determined by the shape of the map around the unstable fixed point, while its chaotic or noisy features play a marginal role. In contrast, in the pushed regime the presence of chaos or noise is more relevant. In particular the front speed decreases when the degree of chaoticity is increased, but it is not straightforward to derive a direct connection between the chaotic properties (e.g. the Lyapunov exponent) and the behaviour of the front. As for the fluctuations of the front position, we observe for the noisy maps that the associated mean square displacement grows in time as t ^1/2 in the pushed case and as t ^1/4 in the pulled one, in agreement with recent findings obtained for continuous models with multiplicative noise. Moreover we show that the same quantity saturates when a chaotic deterministic dynamics is considered for both pushed and pulled regimes. Received 17 July 2001     

Evidence for a compressed fluid phase of Xe clusters

We report on the formation and detection of a compressed fluid phase of Xe clusters in as- implanted Si, at room temperature. The simultaneous structural characterization of the Xe clusters and of the Si matrix was performed by X-ray diffraction at grazing incidence coupled with two-dimensional detection; in both cases, the nearest-neighbor distance and the coordination were obtained. In order to investigate the early stage of the atomic inclusion and the cluster segregation, the average compression and size of Xe fluid clusters within the amorphous Si matrix were explained within the simple Hard Sphere model. Received 4 September 2000 and Received in final form 13 December 2000     

Random inelasticity and velocity fluctuations in a driven granular gas

We analyze the deviations from Maxwell-Boltzmann statistics found in recent experiments studying velocity distributions in two-dimensional granular gases driven into a non-equilibrium stationary state by a strong vertical vibration. We show that in its simplest version, the “stochastic thermostat” model of heated inelastic hard spheres, contrary to what has been hitherto stated, is incompatible with the experimental data, although predicting a reminiscent high-velocity stretched-exponential behavior with an exponent 3/2. The experimental observations lead to refine a recently proposed random restitution coefficient model. Very good agreement is then found with experimental velocity distributions within this framework, which appears self-consistent and further provides relevant probes to investigate the universality of the velocity statistics. Received: 27 May 2002 / Accepted: 6 May 2003 / Published online: 27 May 2003 RID="a" ID="a"e-mail: Alain.Barrat@th.u-psud.fr