Reference priors for exponential families with simple quadratic variance function

Reference analysis is one of the most successful general methods to derive noninformative prior distributions. In practice, however, reference priors are often difficult to obtain. Recently developed theory for conditionally reducible natural exponential families identifies an attractive reparameterization which allows one, among other things, to construct an enriched conjugate prior. In this paper, under the assumption that the variance function is simple quadratic, the order-invariant group reference prior for the above parameter is found. Furthermore, group reference priors for the mean- and natural parameter of the families are obtained. A brief discussion of the frequentist coverage properties is also presented. The theory is illustrated for the multinomial and negative-multinomial family. Posterior computations are especially straightforward due to the fact that the resulting reference distributions belong to the corresponding enriched conjugate family. A substantive application of the theory relates to the construction of reference priors for the Bayesian analysis of two-way contingency tables with respect to two alternative parameterizations.     

Crossover Critical Behavior in the Three-Dimensional Ising Model

The character of critical behavior in physical systems depends on the range of interactions. In the limit of infinite range of the interactions, systems will exhibit mean-field critical behavior, i.e., critical behavior not affected by fluctuations of the order parameter. If the interaction range is finite, the critical behavior asymptotically close to the critical point is determined by fluctuations and the actual critical behavior depends on the particular universality class. A variety of systems, including fluids and anisotropic ferromagnets, belongs to the three-dimensional Ising universality class. Recent numerical studies of Ising models with different interaction ranges have revealed a spectacular crossover between the asymptotic fluctuation-induced critical behavior and mean-field-type critical behavior. In this work, we compare these numerical results with a crossover Landau model based on renormalization-group matching. For this purpose we consider an application of the crossover Landau model to the three-dimensional Ising model without fitting to any adjustable parameters. The crossover behavior of the critical susceptibility and of the order parameter is analyzed over a broad range (ten orders) of the scaled distance to the critical temperature. The dependence of the coupling constant on the interaction range, governing the crossover critical behavior, is discussed.     

Computation of the eigenvalues of Sturm-Liouville problems with parameter dependent boundary conditions using the regularized sampling method

The purpose in this paper is to compute the eigenvalues of Sturm-Liouville problems with quite general separated boundary conditions nonlinear in the eigenvalue parameter using the regularized sampling method, an improvement on the method based on Shannon sampling theory, which does not involve any multiple integration and provides higher order estimates of the eigenvalues at a very low cost. A few examples shall be presented to illustrate the power of the method and a comparison made with the the exact eigenvalues obtained as squares of the zeros of the exact characteristic functions.

     

Sequential Estimation of the Parameters in a Trigonometric Regression Model with the Gaussian Coloured Noise

This paper considers the estimation problem for a trigonometric regression model with the noise specified by the Ornstein–Uhlenbeck process with unknown parameter. We propose a sequential procedure which ensures a prescribed mean square precision uniformly in the nuisance parameter. The asymptotic behaviour of the procedure duration mean has been studied. This revised version was published online in August 2006 with corrections to the Cover Date.     

Novel optical film sensor design based on p-polarized reflectance

A new scheme of optical film sensor is presented. The sensor is based on p-polarized reflectance, consisting of a sensing coated substrate, is easily optimized for maximum sensitivity in different applications. The resolutions of refractive index n_f, extinction coefficient k_f and thickness h_f of the sensitive films are predicted to be 10⁻⁷, 10⁻⁵ and 10⁻³ nm, respectively. Experimentally, we selected the sol–gel derived SnO₂ films as gas-sensitive films and conducted preliminary gas-sensing test. The results indicate that novel optical film sensor scheme has higher sensitivity, and the detection sensitivity is available to 10⁻¹ ppm on the condition of optimum optical parameters and incident angle.     

A Generalized Markov Sampler

A recent development of the Markov chain Monte Carlo (MCMC) technique is the emergence of MCMC samplers that allow transitions between different models. Such samplers make possible a range of computational tasks involving models, including model selection, model evaluation, model averaging and hypothesis testing. An example of this type of sampler is the reversible jump MCMC sampler, which is a generalization of the Metropolis–Hastings algorithm. Here, we present a new MCMC sampler of this type. The new sampler is a generalization of the Gibbs sampler, but somewhat surprisingly, it also turns out to encompass as particular cases all of the well-known MCMC samplers, including those of Metropolis, Barker, and Hastings. Moreover, the new sampler generalizes the reversible jump MCMC. It therefore appears to be a very general framework for MCMC sampling. This paper describes the new sampler and illustrates its use in three applications in Computational Biology, specifically determination of consensus sequences, phylogenetic inference and delineation of isochores via multiple change-point analysis.     

理想流体一维绝热稳定流动能量方程推导

对理想流体一维绝热稳定流动作了讨论．对一般理想流体(可压缩或不可压缩)，用两种方法推出了能量方程．在此基础上，讨论了流动可逆性及不同流体性质对流动参数的影响．     

Robust H∞ control for linear Markovian jump systems with unknown nonlinearities

This paper studies the problem of stochastic stability and disturbance attenuation for a class of linear continuous-time uncertain systems with Markovian jumping parameters. The uncertainties are assumed to be nonlinear and state, control and external disturbance dependent. A sufficient condition is provided to solve the above problem. An H_∞ controller is designed such that the resulting closed-loop system is stochastically stable and has a disturbance attenuation γ for all admissible uncertainties. It is shown that the control law is in terms of the solutions of a set of coupled Riccati inequalities. A numerical example is included to demonstrate the potential of the proposed technique.     

关于三维Minkowski空间中直线汇的一些注记

本文研究了三维Minkowski空间中直线汇的一些性质,特别是关于类时线汇的性质．讨论了线汇基本元素的存在性,并证明了关于三维Minkowski空间中类时线汇的配分参数的一个结果,推广了苏步青1927年的—个成果．     

Collective bounce-off phenomenon in139La+Ag(Br) reaction at 1.1 A GeV energy

Out of a total statistics of 896¹³⁹La+Ag(Br) interactions, 128 interactions having multiplicity of target fragments (Z⩾1)⩾8 and projectile fragments (Z⩾2)⩾4 have been selected. They correspond to quasi-peripheral interactions. Azimuthal angle correlation between sources of target fragments (TFs) and projectile fragments (PFs) shows the existence of bounce-off effect. Using data of La+Ag(Br) and⁸⁴Kr+Ag(Br) reactions it is shown that individual helium [Z=2, PFs] and heavier fragment [Z⩾3, PFs] show different emission characteristics. Further, a two prong correlation functionT(Φ _ij ) plotted for heavier fragments and helium fragments separately, indicates the possibility of existence of different physical conditions. This observation is supported by the different momentum widths of helium fragments and heavier fragments. From the momentum width data of Kr+Ag(Br) reactions normalized density comes out to be ≈4.7. Using quasi-elastic kinematics for the bounce-off nuclei, the excitation energy has been computed from the experimental data of flow angles. The strength of bounce-off seems to decrease with the increase of excitation energy or temperature.