Nonparametric Estimation of Regression Functions in Point Process Models

We prove that the empirical L ²-risk minimizing estimator over some general type of sieve classes is universally, strongly consistent for the regression function in a class of point process models of Poissonian type (random sampling processes). The universal consistency result needs weak assumptions on the underlying distributions and regression functions. It applies in particular to neural net classes and to radial basis function nets. For the estimation of the intensity functions of a Poisson process a similar technique yields consistency of the sieved maximum likelihood estimator for some general sieve classes. This revised version was published online in August 2006 with corrections to the Cover Date.     

多维GARCH模型的半参数有效估计

多元GARCH模型的估计一般采用拟极大似然法(quasi maximum likehood),对于这种方法估计的相合性及渐近正态性已经被很多学者证实,然而对于新息列的分布不是多元正态时,这种估计的有效性还没人研究,本文从拟极大似然估计得到的参数相合估计入手,提出用非参数方法估计多元新息列的分布.     

Development of the Monte Carlo model CASCADE-2004 of high-energy nuclear interactions

The Monte Carlo code CASCADE for the calculation of inelastic hadron- and nucleus-nucleus interactions at energies from several tens of MeV up to several tens of GeV and for modelling of nuclear-physical processes accompanying the transport of particles and nuclei in matter is improved by considering a more detailed model of decay of highly excited residual (after-intranuclear-cascade) nuclei. Results of calculations are in good agreement with experiment. However, there are some deviations for light-isotope production, which prompt the necessity of developing more correct models of evaporation and strong asymmetric high-energy fission.     

0–1 equivalent model representations for air traffic management

We present two equivalent representations of a 0–1 model for the Air Traffic Flow Management Problem. We show that the LP relaxation of the second representation is tighter than the LP relaxation of the first one. As a direct implication, the Branch-and-Cut effort can be reduced. Computational results are reported on a set of large-scale instances. The preparation of this paper is partially supported by DGICYT through grant PB95-0407     

Dynamical properties of low-dimensional and spin-Peierls systems

Properties of low-dimensional spin-Peierls systems are described by using a one-dimensional S =1/2 antiferromagnetic Heisenberg chain linearly coupled to a single phonon mode of wave vector (whose contribution is expected to be dominant). By exact diagonalizations of small rings with up to 24 sites supplemented by a finite size scaling analysis, static and dynamical properties are investigated. Numerical evidences are given for a spontaneous discrete symmetry breaking towards a spin gapped phase with a frozen lattice dimerization. Special emphasis is put on the comparative study of the two inorganic spin-Peierls compounds CuGeO₃ and NaV₂O₅ and the model parameters are determined from a fit of the experimental spin gaps. We predict that the spin-phonon coupling is 2 or 3 times larger in NaV₂O₅ than in CuGeO₃. Inelastic neutron scattering spectra are calculated and similar results are found in the single phonon mode approximation and in the model including a static dimerization. In particular, the magnon S =1 branch is clearly separated from the continuum of triplet excitations by a finite gap. Received: 30 July 1997 / Revised: 16 September 1997 / Accepted: 10 October 1997     

Discrete symmetry groups of vertex models in statistical mechanics

We analyze discrete symmetry groups of vertex models in lattice statistical mechanics represented as groups of birational transformations. They can be seen as generated by involutions corresponding respectively to two kinds of transformations onq×q matrices: the inversion of theq×q matrix and an (involutive) permutation of the entries of the matrix. We show that the analysis of the factorizations of the iterations of these transformations is a precious tool in the study of lattice models in statistical mechanics. This approach enables one to analyze two-dimensionalq ⁴-state vertex models as simply as three-dimensional vertex models, or higher-dimensional vertex models. Various examples of birational symmetries of vertex models are analyzed. A particular emphasis is devoted to a three-dimensional vertex model, the 64-state cubic vertex model, which exhibits a polynomial growth of the complexity of the calculations. A subcase of this general model is seen to yield integrable recursion relations. We also concentrate on a specific two-dimensional vertex model to see how the generic exponential growth of the calculations reduces to a polynomial growth when the model becomes Yang-Baxter integrable. It is also underlined that a polynomial growth of the complexity of these iterations can occur even for transformations yielding algebraic surfaces, or higher-dimensional algebraic varieties.     

Nucleon Spin Structure Functions in the Resonance Region and the Duality

We discuss the nucleon spin structure function gl and the difference between the proton and neutrontargets gp1 - gn1 , based on quark model calculation. Quark-hadron duality for the nucleon spin structure function is alsoanalyzed. Effects of the △(1232) and Roper P11(1440) resonances on the spin structure function and on the differencegn1 - gn1 are mentioned. The results of different models for the Roper resonance are also addressed.     

On the formation of sharp transition layers in two-dimensional reaction-diffusion models

For a singularly perturbed parabolic equation in two dimensions, the formation of a solution with a sharp transition layer from a sufficiently general initial function is considered. An asymptotic analysis is used to estimate the time required for the formation of a contrast structure. Numerical results are presented.