Inferential aspects of the zero-inflated Poisson INAR(1) process

A first-order INteger-valued AutoRegressive (INAR) process with zero-inflated Poisson distributed innovations was proposed by Jazi, Jones and Lai (2012) [First-order integer valued AR processes with zero inflated Poisson innovations. Journal of Time Series Analysis. 33, 954–963.], which is able for dealing with zero-inflated/deflated count time series data. The inferential aspects of this model were not well explored by the authors, only a conditional maximum likelihood approach was briefly discussed. In this paper, we explore the inferential aspects of this zero-inflated Poisson INAR(1) process. We propose parameter estimation through Two-Step Conditional Least Squares and Yule–Walker methods. The asymptotic properties of the estimators are provided. Simulation results about the finite-sample behavior of both estimation methods and comparisons with the conditional maximum likelihood approach are presented under correct model specification and misspecification. Two empirical applications to real data sets are considered in order to illustrate the usefulness of the proposed methodology in practical situations.     

Risk aggregation based on the Poisson INAR(1) process with periodic structure

In this paper, we consider a risk model by introducing a temporal dependence between the claim numbers under periodic environment, which generalizes several discrete-time risk models. The model proposed is based on the Poisson INAR(1) process with periodic structure. We study the moment-generating function of the aggregate claims. The distribution of the aggregate claims is discussed when the individual claim size is exponentially distributed.     

Validation tests for the innovation distribution in INAR time series models

Goodness-of-fit tests are proposed for the innovation distribution in INAR models. The test statistics incorporate the joint probability generating function of the observations. Special emphasis is given to the INAR(1) model and particular instances of the procedures which involve innovations from the general family of Poisson stopped-sum distributions. A Monte Carlo power study of a bootstrap version of the test statistic is included as well as a real data example. Generalizations of the proposed methods are also discussed.     

Controlling jumps in correlated processes of Poisson counts

Processes of autocorrelated Poisson counts can often be modelled by a Poisson INAR(1) model, which proved to apply well to typical tasks of SPC. Statistical properties of this model are briefly reviewed. Based on these properties, we propose a new control chart: the combined jumps chart. It monitors the counts and jumps of a Poisson INAR(1) process simultaneously. As the bivariate process of counts and jumps is a homogeneous Markov chain, average run lengths (ARLs) can be computed exactly with the well‐known Markov chain approach. Based on an investigation of such ARLs, we derive design recommendations and show that a properly designed chart can be applied nearly universally. This is also demonstrated by a real‐data example from the insurance field. Copyright © 2008 John Wiley & Sons, Ltd.     

Quantile Regression for Thinning-based INAR(1) Models of Time Series of Counts

In this paper,we develop the quantile regression(QR)estimation for the first-order integer-valued autoregressive(INAR(1))models by defining the smoothing INAR(1)process.Jittering method is used to derive the QR estimators for the autoregressive coefficient and the quantile of innovations.The consistency and asymptotic normality of the proposed estimators are established.The performances of the proposed estimation procedures are evaluated by Monte Carlo simulations.The results show that the proposed procedures perform well for simulations and a real data application.     

Stieltjes representation of the 3D Bruggeman effective medium and Padé approximation

The paper deals with Bruggeman effective medium approximation (EMA) which is often used to model effective complex permittivity of a two-phase composite. We derive the Stieltjes integral representation of the 3D Bruggeman effective medium and use constrained Padé approximation method introduced in [39] to numerically reconstruct the spectral density function in this representation from the effective complex permittivity known in a range of frequencies. The problem of reconstruction of the Stieltjes integral representation arises in inverse homogenization problem where information about the spectral function recovered from the effective properties of the composite, is used to characterize its geometric structure. We present two different proofs of the Stieltjes analytical representation for the effective complex permittivity in the 3D Bruggeman effective medium model: one proof is based on direct calculation, the other one is the derivation of the representation using Stieltjes inversion formula. We show that the continuous spectral density in the integral representation for the Bruggeman EMA model can be efficiently approximated by a rational function. A rational approximation of the spectral density is obtained from the solution of a constrained minimization problem followed by the partial fractions decomposition. We show results of numerical rational approximation of Bruggeman continuous spectral density and use these results for estimation of fractions of components in a composite from simulated effective permittivity of the medium. The volume fractions of the constituents in the composite calculated from the recovered spectral function show good agreement between theoretical and predicted values.     

Bayesian Estimation for the Order of INAR(q) Mo del

In this paper, we consider the problem of determining the order of INAR(q) model on the basis of the Bayesian estimation theory. The Bayesian es-timator for the order is given with respect to a squared-error loss function. The consistency of the estimator is discussed. The results of a simulation study for the estimation method are presented.     

Spectral methods for bivariate Markov processes with diffusion and discrete components and a variant of the Wright–Fisher model

The aim of this paper is to study differential and spectral properties of the infinitesimal operator of two dimensional Markov processes with diffusion and discrete components. The infinitesimal operator is now a second-order differential operator with matrix-valued coefficients, from which we can derive backward and forward equations, a spectral representation of the probability density, study recurrence of the process and the corresponding invariant distribution. All these results are applied to an example coming from group representation theory which can be viewed as a variant of the Wright–Fisher model involving only mutation effects.     

Integer Valued Autoegressive Process with Katz Arrivals and Its Application in Predicting the Count of the Cases of Respiratory Disease

The traditional PAR process (Poisson autoregressive process) assumes that the arrival process is the equi-dispersed Poisson process, with its mean being equal to its variance. Whereas the arrival process in the real DGP (data generating process) could either be over-dispersed, with variance being greater than the mean, or under-dispersed, with variance being less than the mean. This paper proposes using the Katz family distributions to model the arrival process in the INAR process (integer valued autoregressive process with Katz arrivals) and deploying Monte Carlo simulations to examine the performance of maximum likelihood (ML) and method of moments (MM) estimators of INAR-Katz model. Finally, we used the INAR-Katz process to model count data of hospital emergency room visits for respiratory disease. The results show that the INAR-Katz model outperforms the Poisson model, PAR(1) model, and has great potential in empirical application.     

EIN NEUER BEWEIS UND EINE VERSCHÄRFUNG FÜR DEN REDUKTIONSTYP ∀∃∀∞(0, 1) MIT EINER ANWENDUNG AUF DIE SPEKTRALE DARSTELLUNG VON PRÄDIKATEN

A simultaneous semantical and syntactical reduction is given for the satisfiability respectively finite satisfiability of first order formulas. We choose ???∞(0, 1) as conservative reduction class and allow only formulas out of ???∞(0, 1) having a simple set theoretical model if they are satisfiable at all. With the same method we get a spectral representation of any ?-ary enumerable respectively coenumerable predicate by a formula out of ???∞(?, 1).     

一类m-KP方程的谱表示及其拟周期解

This paper gives the spectral representation of a class of (2 1)-dimensional mod- ified Kadomtsev-Petviashvili(m-KP) equation with a constant parameter.Its quasi-periodic solution is obtained in terms of Riemann theta functions.     

The problem of harmonic analysis on the infinite-dimensional unitary group

The goal of harmonic analysis on a (noncommutative) group is to decompose the most “natural” unitary representations of this group (like the regular representation) on irreducible ones. The infinite-dimensional unitary group U(∞) is one of the basic examples of “big” groups whose irreducible representations depend on infinitely many parameters. Our aim is to explain what the harmonic analysis on U(∞) consists of.We deal with unitary representations of a reasonable class, which are in 1-1 correspondence with characters (central, positive definite, normalized functions on U(∞)). The decomposition of any representation of this class is described by a probability measure (called spectral measure) on the space of indecomposable characters. The indecomposable characters were found by Dan Voiculescu in 1976.The main result of the present paper consists in explicitly constructing a 4-parameter family of “natural” representations and computing their characters. We view these representations as a substitute of the nonexisting regular representation of U(∞). We state the problem of harmonic analysis on U(∞) as the problem of computing the spectral measures for these “natural” representations. A solution to this problem is given in the next paper (Harmonic analysis on the infinite-dimensional unitary group and determinantal point processes, math/0109194, to appear in Ann. Math.), joint with Alexei Borodin.We also prove a few auxiliary general results. In particular, it is proved that the spectral measure of any character of U(∞) can be approximated by a sequence of (discrete) spectral measures for the restrictions of the character to the compact unitary groups U(N). This fact is a starting point for computing spectral measures.     

THE STRONG LAW OF LARGE NUMBER AND PARAMETER ESTIMATION OF ONE CLASS OF NON-NEGATIVE INTEGER-VALUED TIME SERIES

1.IntroductionLet{xt}beatimeseriesdefinedonprobabilityspace{fi,F,P),satisfyingwhere(1)p(t,s,j)isarandomfieldwithValueIor0,t,8EI={0,11,f2,'.},jEN={1,2,'.},andsatisfies(i)p{p(t,s,j)=0}=1whens>t;(1.2)(if)ac(tl,sl,jl)andp(tz,sa,j~)aremutuallyindependentwhenif/iZor81/s2;(ill)FOrfords,j,p(t,s,i)fort2sisanonhomogeneousMarkovChainwithtransitionprobabilityandinitialdistributiont=0T=0(2){Mt}isani.i.d.non-negativeinteger-valuedtimeseriesandindependentofn(t,sli).Inthefollowing,weassumeEMtz相似文献   

A generalization of the Fourier transform and its application to spectral analysis of chirp-like signals

We show that the de Branges theory provides a useful generalization of the Fourier transform (FT). The formulation is quite rich in that by selecting the appropriate parametrization, one can obtain spectral representation for a number of important cases. We demonstrate two such cases in this paper: the finite sum of elementary chirp-like signals, and a decaying chirp using Bessel functions. We show that when defined in the framework of de Branges spaces, these cases admit a representation very much similar to the spectral representation of a finite sum of sinusoids for the usual FT.     

Absolutely continuous spectrum of a nondissipative operator and the functional model. I

In the paper one investigates nondissipative operators L in a Hilbert space; for them one constructs a model representation similar to the Nagy-Foias model for dissipative operators. In this representation one succeeds to calculate the action of the resolvent of a nondissipative operator on selected subspaces. This allows us to relinquish the consideration of the -self-adjoint dilation of the operator, whose spectral representation involves considerable difficulties. Isolated results are new also for the dissipative case which is not excluded. In part I one considers the “triangular” factorization of the characteristic funcion of the operator L and one carries out the proof of the fundamental theorem which gives a formula for the calculation of (L-No)⁻¹(J_mN_oO T_mN_o<O) on selected subspaces. The applications of this theorem to the spectral analysis on the absolutely continuous spectrum and to the problems of linear similitude for nondissipative operators are considered in part II. Translated from Zapiski Nauchnykh Seminarov Leningradskogo Otdeleniya Matematicheskogo Ihstituta im. V. A. Steklova AN SSSR, Vol. 65, pp. 90–102, 1976.     

The SLEX Model of a Non-Stationary Random Process

We propose a new model for non-stationary random processes to represent time series with a time-varying spectral structure. Our SLEX model can be considered as a discrete time-dependent Cramér spectral representation. It is based on the so-called Smooth Localized complex EXponential basis functions which are orthogonal and localized in both time and frequency domains. Our model delivers a finite sample size representation of a SLEX process having a SLEX spectrum which is piecewise constant over time segments. In addition, we embed it into a sequence of models with a limit spectrum, a smoothly in time varying evolutionary spectrum. Hence, we develop the SLEX model parallel to the Dahlhaus (1997, Ann. Statist., 25, 1–37) model of local stationarity, and we show that the two models are asymptotically mean square equivalent. Moreover, to define both the growing complexity of our model sequence and the regularity of the SLEX spectrum we use a wavelet expansion of the spectrum over time. Finally, we develop theory on how to estimate the spectral quantities, and we briefly discuss how to form inference based on resampling (bootstrapping) made possible by the special structure of the SLEX model which allows for simple synthesis of non-stationary processes.     

Measures of Non-compactness of Operators on Banach Lattices

[Indag. Math.(N.S.) 2(2) (1991), 149–158; Uspehi Mat. Nauk 27(1(163)) (1972), 81–146] used representation spaces to study measures of non-compactness and spectral radii of operators on Banach lattices. In this paper, we develop representation spaces based on the nonstandard hull construction (which is equivalent to the ultrapower construction). As a particular application, we present a simple proof and some extensions of the main result of [J. Funct. Anal. 78(1) (1988), 31–55] on the monotonicity of the measure of non-compactness and the spectral radius of AM-compact operators. We also use the representation spaces to characterize d-convergence and discuss the relationship between d-convergence and the measures of non-compactness.     

On the structure of the essential spectrum of a model many-body Hamiltonian

In this paper, we study the essential spectrum of a model lattice Hamiltonian describing a system with fluctuating number of particles (0 ≤ n ≤ 2) in the quasimomentum representation. The spectral properties are described in terms of the boundary values of a function of a complex variable, whose meaning is that of the kernel of the Schur complement $$ H_{11} - z - H_{12} \left( {H_{22} - z} \right)^{ - 1} H_{12}^ * . $$      

随机变量二次型的协方差在混合效应模型中的应用

本文提出方差分量ANOVA估计的一种改进方法, 证明了对于一般的方差分量模型, 只要方差分量的ANOVA估计存在就可以通过此方法给出其改进形式, 并且在均方误差意义下优于ANOVA估计. 特别地, 对于单向分类随机效应模型, Kelly和Mathew^[1]对ANOVA估计的改进就是我们提出的改进方法的特殊形式, 这也给出了此类改进估计在均方误差意义下优于ANOVA估计的另一种合理的解释. 同时, 本文又将此思想应用到对谱分解估计的改进上. 本文应用协方差的简单性质证明了对带有一个随机效应的方差分量模型, 当随机效应的协方差阵只有一个非零特征值时, 随机效应方差分量谱分解估计在均方误差意义下总是优于ANOVA估计. 本文最后将第三节的结论推广到广义谱分解估计下, 同时给出广义谱分解估计待定系数的一个合理的取值.     

Spectral Theory for Discrete Laplacians

We give the spectral representation for a class of selfadjoint discrete graph Laplacians Δ, with Δ depending on a chosen graph G and a conductance function c defined on the edges of G. We show that the spectral representations for Δ fall in two model classes, (1) tree-graphs with N-adic branching laws, and (2) lattice graphs. We show that the spectral theory of the first class may be computed with the use of rank-one perturbations of the real part of the unilateral shift, while the second is analogously built up with the use of the bilateral shift. We further analyze the effect on spectra of the conductance function c: How the spectral representation of Δ depends on c.