Asymptotic distribution of Wishart matrix for block-wise dispersion of population eigenvalues

This paper deals with the asymptotic distribution of Wishart matrix and its application to the estimation of the population matrix parameter when the population eigenvalues are block-wise infinitely dispersed. We show that the appropriately normalized eigenvectors and eigenvalues asymptotically generate two Wishart matrices and one normally distributed random matrix, which are mutually independent. For a family of orthogonally equivariant estimators, we calculate the asymptotic risks with respect to the entropy or the quadratic loss function and derive the asymptotically best estimator among the family. We numerically show (1) the convergence in both the distributions and the risks are quick enough for a practical use, (2) the asymptotically best estimator is robust against the deviation of the population eigenvalues from the block-wise infinite dispersion.     

A class of PP l-statistics for tests

Applying projection pursuit (PP) techniques, this paper takes L-statistics as projection indices to construct a class of statistics for multivariate tests. The asymptotic distributions of the test statistics under null hypothesis are derived theoretically. Some simple applications to location and dispersion problems are given.This project is supported by the National Natural Science Foundation of China.     

On Nonpreemptive LCFS Scheduling with Deadlines

We investigate some real time behaviour of a (discrete time) single server system with nonpreemptive LCFS task scheduling. The main results deal with the probability distribution of a random variable SRD(T), which describes the time the system operates without any violation of a fixed task service time deadline T. A tree approach, similar to those already used for the derivation of the same quantities for other scheduling disciplines (e.g., FCFS) is suitable here again, establishing the power of such techniques once more. Relying on a simple general probability model, asymptotic formulas concerning all moments of SRD(T) are determined; for example, the expectation of SRD(T) is proved to grow exponentially in T, i.e., E[SRD(T)] ∼ CT^3/2ρ^T for some ρ > 1. Our computations rely on a multivariate (asymptotic) coefficient extraction technique which we call asymptotic separation.     

Moments for a multivariate linear model with an application to the growth curve model

An extended growth curve model is considered which, among other things, is useful when linear restrictions exist on the mean in the ordinary growth curve model. The maximum likelihood estimators consist of complicated stochastic expressions. It is shown how, by the aid of fairly elementary calculations, the dispersion matrix for the estimator of the mean and the expectation of the estimated dispersion matrix are obtained. Results for Wishart, inverted Wishart, and inverse beta variables are utilized. Additionally, some asymptotic results are presented.     

Jackknifing in generalized linear models

In a generalized linear model, the jackknife estimator of the asymptotic covariance matrix of the maximum likelihood estimator is shown to be consistent. The corresponding jackknife studentized statistic is asymptotically normal. In addition, these results remain true even if there exist unequal dispersion parameters in the model. On the other hand, the variance estimator and the studentized statistic based on the standard method (substitution and linearization) do not enjoy this robustness property against the presence of unequal dispersion parameters.This research was supported by an Operating Grant from the Natural Science and Engineering Research Council of Canada.     

协方差矩阵的齐性检验

基于ＰＰ技术、Ｂｏｏｔｓｔｒａｐ方法和数论方法,对于ｋ个总体协方差矩阵相等的检验,给出了ＰＰ型检验统计量,并讨论了它的渐近分布和Ｂｏｏｔｓｔｒａｐ逼近,最后给出了一些实际模拟结果。     

A finite-element approach in order to avoid ill-conditioning in thin-sheet problems in frequency domain—Application to magneto-quasistatics

In this work, a new thin-sheet approach in the finite-element method is derived. The focus is on the condition number of the system matrix, namely, to keep this measure preferably independent of the thickness of the sheet. Constant sheet elements are used for the tangential variation in the sheet. However, the information about the discontinuity in normal direction is incorporated into the basis functions of the volume elements that are connected to the sheet elements. The determination of the normal variation can be reduced to a 1D problem which can be solved analytically. No double layers or global asymptotic expansions are required. The advantages with respect to the condition number of the system matrix are shown for a magneto-quasistatic test scenario.     

Multivariate sequential point estimation

For a multivariate normal distribution with unknown mean vector and unknown dispersion matrix, a sequential procedure for estimating the unknown mean vector is suggested. The procedure is shown to be asymptotically “risk efficient” in the sense of Starr (Ann. Math. Statist. (1966), 1173–1185), and the asymptotic order of the “regret” (see Starr and Woodroofe, Proc. Nat. Acad. Sci. 63 (1969), 285–288) is given. Moderate sample behaviour of the procedure using Monte-Carlo techniques is also studied. Finally, the asymptotic normality of the stopping time is proved.     

Asymptotic distributions of robust shape matrices and scales

It has been frequently observed in the literature that many multivariate statistical methods require the covariance or dispersion matrix Σ of an elliptical distribution only up to some scaling constant. If the topic of interest is not the scale but only the shape of the elliptical distribution, it is not meaningful to focus on the asymptotic distribution of an estimator for Σ or another matrix Γ∝Σ. In the present work, robust estimators for the shape matrix and the associated scale are investigated. Explicit expressions for their joint asymptotic distributions are derived. It turns out that if the joint asymptotic distribution is normal, the estimators presented are asymptotically independent for one and only one specific choice of the scale function. If it is non-normal (this holds for example if the estimators for the shape matrix and scale are based on the minimum volume ellipsoid estimator) only the scale function presented leads to asymptotically uncorrelated estimators. This is a generalization of a result obtained by Paindaveine [D. Paindaveine, A canonical definition of shape, Statistics and Probability Letters 78 (2008) 2240-2247] in the context of local asymptotic normality theory.     

Influence Function and Efficiency of the Minimum Covariance Determinant Scatter Matrix Estimator

The minimum covariance determinant (MCD) scatter estimator is a highly robust estimator for the dispersion matrix of a multivariate, elliptically symmetric distribution. It is relatively fast to compute and intuitively appealing. In this note we derive its influence function and compute the asymptotic variances of its elements. A comparison with the one step reweighted MCD and with S-estimators is made. Also finite-sample results are reported.     

The dissipation of nonlinear dispersive waves: the case of asymptotically weak nonlinearity

It is proved herein that certain smooth, global solutions of a class of quasi-linear, dissipative wave equations have precisely the same leading order, long-time, asymptotic behavior as the solutions with the same initial data of the corresponding linearized equations. The solutions of the nonlinear equations are shown to be asymptotically self-similar with explicitly determined profiles. The equations considered have homogeneous nonlinearities and homogeneous dispersive and dissipative symbols. By relating these degrees of homogeneous to the leading order asymptotic behavior of the Fourier transform of the initial data near k= 0, different classes of long-time asymptotic behavior are characterized. These results cover the case where dissipation is not asymptotically negligible in comparison with dispersion, and where nonlinear effect are asymptotically negligible in comparison with linear effect, i.e., dissipation and dispersion. They always hold for solutions with "small" initial data. In most circumstances however a new a priori bound on certain negative homogeneous Sobolev norms of solutions is obtained, which implies that any solution, even one which is initially "large" will eventually satisfy the smallness condition, and hence will have the above described asymptotic behavior     

On degenerate multi-dimensional dispersion laws

Degeneration of multi-dimensional dispersion laws analytic in the vicinity of zero is investigated under the assumption that the corresponding degeneracy function satisfies a definite condition (Eq. (3)). It is proved that only two-dimensional dispersion laws ω(p, q), whose asymptotic behavior at small p and q has some definite form (Eq. (28)), can be degenerate with respect to the decay process 1 → 2. It is shown that the corresponding degeneracy function is unique and its asymptotic behavior is found. Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 112, No. 1, pp. 124–131.     

Asymptotic behaviour and level-curve structure in plane subsonic potential flows

The problem of constructing asymptotic forms at infinity and the problem of determining the structure of the isoclines and isobars are considered for uniform plane subsonic potential flow, horizontal at infinity, around a large class of bodies. It is shown that these problems are intimately related. In fact, the construction of a solution in the neighbourhood of the point at infinity (PAI) reduces to (i) selecting a “correct” transformation of the physical plane (PP) onto an auxiliary plane (AP), under which the PAI of the PP goes into the origin of the AP and the gas dynamic equations at the origin of the AP reduce to a Cauchy-Riemann system; (ii) finding the number of isoclines that pass through the PAI and determining the inclinations of these isoclines. With this approach, the construction of asymptotic laws and the investigation of the level curve structure in the neighbourhood of the PAI have much in common with the analogous problem in the neighbourhood of an arbitrary point of the flow at a finite distance from the body. Asymptotic forms are constructed for two cases: lift-creating flow and symmetric flow around a body. The constant factors occurring in the asymptotic formulae are expressed in terms of the physical or geometrical parameters of the problems under consideration.     

PP 检验的渐近功效

投影寻踪(Projection Pursnit,简称 PP)是一种处理高维数据的统计方法,近年来借助于计算机的发展,它的理论得到了迅速的发展.假如(?)为一能反映统计性质的指标,对每个 p 维方向α,计算出(?)(α~(?)X_1,…,α~(?)X_(?)),从中找出使(?)(α~(?)?X_1,…,α~(?)X_(?))最大的方向α_0,通过研究数据α_0~(?)X_1,…,α_0~(?)X_(?) 的性质,来了解原数据 X_1,…,X_(?)的性质,这就是所谓数值 PP 的基本想法.如果用一维检验或估计统计量(?)作指标,则可得到多维检验或估计(?)(?)(α~(?)X_1,…,α~(?)X_(?)).见文献[2,3,5—7].在上述文章中这些检验和     

PP multivariate random weighting method

According to the Projection Pursuit (PP) method and the random weighting method, we propose a PP random weighting method, and set up the asymptotic distribution theory and strong limit theorem of PP random weighting empirical process. Applying this method, we obtain two kinds of goodness-of-fit test for a multivariate distribution function, i.e., we get the random weighting approximations of PP Kolmogorov Smirnov statistics (PPKS) and PP Smirnov Cramér Von Mises statistics (PPSC), we prove that the asymptotic distribution of PPKS and PPSC are the same as those of their respective random weighting approximations.Supported by the National Natural Science Foundation of China.     

Analytical solutions to detect the scheme dispersion for the coupled nonlinear equations

It is shown, how even particular traveling wave asymptotic solution may describe the defects on the shock wave profile caused by the dispersion features of the numerical scheme of the coupled nonlinear gas dynamics equations. For this purpose the coupled nonlinear partial differential equations or the so-called differential approximation of the scheme, are obtained, and a simplification of the method of differential approximation is suggested to obtain the desired asymptotic solution. The solution is used to study the roles of artificial viscosity and the refinement of the mesh for the suppression of the dispersion of the scheme.     

Asymptotic expansions for the pivots using log-likelihood derivatives with an application in item response theory

Asymptotic expansions of the distributions of the pivotal statistics involving log-likelihood derivatives under possible model misspecification are derived using the asymptotic cumulants up to the fourth-order and the higher-order asymptotic variance. The pivots dealt with are the studentized ones by the estimated expected information, the negative Hessian matrix, the sum of products of gradient vectors, and the so-called sandwich estimator. It is shown that the first three asymptotic cumulants are the same over the pivots under correct model specification with a general condition of the equalities. An application is given in item response theory, where the observed information is usually used rather than the estimated expected one.     

Boundary conditions and boundary layers for a multi-dimensional relaxation model

In this paper we study the linearized relaxation model of Katsoulakis and Tzavaras in a half-space with arbitrary space dimension n?1. Our main interest is to establish the asymptotic equivalence of the relaxation system and its corresponding multi-dimensional equilibrium conservation law. We identify and rigorously justify a necessary and sufficient condition (which we refer to as stiff Kreiss condition, or SKC in short) on the boundary condition to guarantee the uniform stability of the initial-boundary value problem of the relaxation system independent of the relaxation rate. The asymptotic convergence and the corresponding boundary layer behavior are studied by Fourier-Laplace transform and a detailed asymptotic analysis. The SKC is shown to be more restrictive than the classical uniform Kreiss condition for all n?1.     

Global asymptotic stability of Lotka–Volterra competition systems with diffusion and time delays

In the Lotka–Volterra competition system with N-competing species if the effect of dispersion and time-delays are both taken into consideration, then the densities of the competing species are governed by a coupled system of reaction–diffusion equations with time-delays. The aim of this paper is to investigate the asymptotic behavior of the time-dependent solution in relation to a positive uniform solution of the corresponding steady-state problem in a bounded domain with Neumann boundary condition, including the existence and uniqueness of a positive steady-state solution. A simple and easily verifiable condition is given to the competing rate constants to ensure the global asymptotic stability of the positive steady-state solution. This result leads to the permanence of the competing system, the instability of the trivial and all forms of semitrivial solutions, and the nonexistence of nonuniform steady-state solution. The condition for the global asymptotic stability is independent of diffusion and time-delays, and the conclusions for the reaction–diffusion system are directly applicable to the corresponding ordinary differential system.     

Consistency and asymptotic normality of profilekernel and backfitting estimators in semiparametric reproductive dispersion nonlinear models

Semiparametric reproductive dispersion nonlinear model (SRDNM) is an extension of nonlinear reproductive dispersion models and semiparametric nonlinear regression models, and includes semiparametric nonlinear model and semiparametric generalized linear model as its special cases. Based on the local kernel estimate of nonparametric component, profile-kernel and backfitting estimators of parameters of interest are proposed in SRDNM, and theoretical comparison of both estimators is also investigated in this paper. Under some regularity conditions, strong consistency and asymptotic normality of two estimators are proved. It is shown that the backfitting method produces a larger asymptotic variance than that for the profile-kernel method. A simulation study and a real example are used to illustrate the proposed methodologies. This work was supported by National Natural Science Foundation of China (Grant Nos. 10561008, 10761011), Natural Science Foundation of Department of Education of Zhejiang Province (Grant No. Y200805073), PhD Special Scientific Research Foundation of Chinese University (Grant No. 20060673002) and Program for New Century Excellent Talents in University (Grant No. NCET-07-0737)