ARCH模型的多变点检验

研究自回归条件异方差(ARCH)模型的多变点检验问题.提出一种拟似然比检验统计量,并在原假设下给出统计量的极限分布.在假设检验过程中得到变点个数的一致估计.数值模拟与实例分析说明了方法的合理性.     

非参数模型均值函数结构变点的Bootstrap检测

本文检测非参数回归模型均值函数结构变点,针对均值函数跃度的长期均值为零时,基于残量的CUSUM统计量对均值函数结构变点检验无效的问题,本文提出了一种基于均值函数的核估计的检验统计量,得到统计量在原假设和备择假设下的极限分布,并构造Bootstrap方法对非参数回归模型均值函数结构变点进行检验,证明了检验和估计的一致性;模拟结果表明本文方法明显优于已有方法。     

长相依序列均值变点检测

研究了长相依序列均值变点检测问题.首先给出变点检验的Ratio统计量其次分别推导了原假设和备择假设下统计量的极限分布,最后用蒙特卡洛方法模拟出检验的临界值,并通过数值模拟和实例分析说明方法的有效性.     

GARCH模型变点的Ratio检验

研究GARCH模型参数变点的Ratio检验.首先构造了基于残量累积平方和的Ratio统计量,推导了原假设下统计量的极限分布,其次采用Monte Carlo方法检验其有效性,最后以数据为例进一步说明该方法的实用性.     

线性模型中M检验原假设分布的随机加权逼近

在线性模型中M-方法可以用于线性假设检验, 其中M检验、Wald检验和Rao的计分型检验是最常用的检验准则. 但是在计算这些检验的临界值时都涉及到未知参数的估计. 在本文中我们利用随机加权的方法来逼近这些检验的原假设分布. 结果表明在原假设和局部对立假设之下随机加权统计量的渐近分布与原检验统计量在原假设之下的渐近分布相同. 因此我们不需要对冗余参数进行估计,利用随机加权的方法就可以得到这些检验的临界值. 而且在局部对立假设之下可以实现对功效的计算. 当取不同的误差分布和不同的随机权时, 我们对本文的方法进行了蒙特卡洛模拟. 结果表明用随机加权方法来逼近原假设分布是非常精确的.     

基于经验欧氏似然的均值单变点检测

均值单变点检测是研究变点问题的基础.论文根据均值单变点模型的特点,构造截断经验欧氏似然比检验函数并给出显式表达.在此基础上,得到了零假设下检验统计量的极限分布为极值分布,给出变点的诊断方法.在有变点的情况下,进一步给出变点位置的估计和其相合性的理论证明.最后通过数值模拟和尼罗河年流量的实证分析说明所提方法的有效性和实用性.     

部分线性变系数模型的Profile Lagrange乘子检验

对于部分线性变系数模型附有约束条件时的估计与检验问题,基于Profile最小二乘方法给出了参数部分以及非参数部分的约束估计并研究了它们的渐近性质,并针对约束条件构造了Profile Lagrange乘子检验统计量,证明了该统计量在原假设下的渐近分布为χ2分布,从而将Lagrange乘子检验方法推广到了半参数模型上.     

相依样本下半参数变系数期望分位数风险度量模型统计推断

本文在α-混合序列假设下,基于半参数变系数模型研究条件期望分位数风险价值(expectile-based value at risk, EVaR)的风险度量.此模型不仅考虑了风险因素的影响,还可以动态描述风险影响及交互效应.同时, EVaR比经典的风险在险价值(quantile-based value at risk, QVaR)具有更直观、更易于计算的良好性质,而且对于资产分布的尾部损失更加敏感,在度量极端风险情形下,相对于QVaR更为有效和方便.本文采用三阶段估计的方法,分别对变系数部分和常系数部分的参数进行估计,并且给出3个阶段中每个估计的相合性和渐近正态性.为了节省计算时间,提高计算效率,本文采用一步估计的算法,减少迭代所需的时间.由于时间序列样本是非独立样本,建立这些统计量的大样本性质时带来了更大的困难.有别于独立同分布的观察数据,本文利用大小块分割方法发展α-混合序列的极限理论,获得了基于金融时间序列数据建立的模型参数和非参数估计的统计渐近性质.在数值模拟中,本文给出3个模型假设下变系数曲线估计和常系数估计的结果,无论是估计的精确度还是估计的稳健性,模拟结果都表明本文所提出的估计方法有优良的性质.实例则展示了本文所提出模型在上证指数的实际应用.     

周期异方差时间序列的季节单位根检验

为检验带异方差的季节时间序列中的单位根,提出基于最小二乘估计的统计量.在原假设下得到检验统计量的极限分布.用M on te C arlo方法计算同方差下的经验分位数并考虑异方差对检验水平的影响.实例分析表明了用该方法检验季节单位根的有效性.     

基于Bootstrap方法的AR(p)模型方差变点的检验

研究了基于AR(p)模型结构变点的检验.原假设条件成立时,证明了残量平方累积和统计量的极限分布仍然是一个标准布朗桥的上确界,并利用bootstrap抽样方法以提高经验势函数值.数值模拟和实例分析都充分说明方法对相依过程结构变点检验的有效性.     

Computational methods of solving the boundary value problems for the loaded differential and Fredholm integro‐differential equations

The article presents a new general solution to a loaded differential equation and describes its properties. Solving a linear boundary value problem for loaded differential equation is reduced to the solving a system of linear algebraic equations with respect to the arbitrary vectors of general solution introduced. The system's coefficients and right sides are computed by solving the Cauchy problems for ordinary differential equations. Algorithms of constructing a new general solution and solving a linear boundary value problem for loaded differential equation are offered. Linear boundary value problem for the Fredholm integro‐differential equation is approximated by the linear boundary value problem for loaded differential equation. A mutual relationship between the qualitative properties of original and approximate problems is obtained, and the estimates for differences between their solutions are given. The paper proposes numerical and approximate methods of solving a linear boundary value problem for the Fredholm integro‐differential equation and examines their convergence, stability, and accuracy.     

Stability in the almost everywhere sense: A linear transfer operator approach

The problem of almost everywhere stability of a nonlinear autonomous ordinary differential equation is studied using a linear transfer operator framework. The infinitesimal generator of a linear transfer operator (Perron-Frobenius) is used to provide stability conditions of an autonomous ordinary differential equation. It is shown that almost everywhere uniform stability of a nonlinear differential equation, is equivalent to the existence of a non-negative solution for a steady state advection type linear partial differential equation. We refer to this non-negative solution, verifying almost everywhere global stability, as Lyapunov density. A numerical method using finite element techniques is used for the computation of Lyapunov density.     

Necessary conditions and a test for the stability of a system of two linear ordinary differential equations of the first order

We use the Riccati equation method for the derivation of necessary conditions and a test for the stability of a system of two linear first-order ordinary differential equations. We consider an example in which our results are compared with the results obtained by the Lyapunov and Bogdanov methods by estimating the norms of solutions via Lozinskii logarithmic norms, and by the freezing method.     

Asymptotic mean-square stability of linear multi-step methods for SODEs

In this article we present results of a linear stability analysis of stochastic linear multi-step methods for stochastic ordinary differential equations. As in deterministic numerical analysis we use a linear time-invariant test equation and study when the numerical approximation shares asymptotic properties in the mean-square sense of the exact solution of that test equation. Sufficient conditions for asymptotic mean-square stability of stochastic linear two-step-Maruyama methods are obtained with the aide of Lyapunov-type functionals. In particular we study the asymptotic mean-square stability of stochastic counterparts of two-step Adams-Bashforth- and Adams-Moulton-methods and the BDF method. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Numerical method of solution to loaded nonlocal boundary value problems for ordinary differential equations

A numerical method is suggested for solving systems of nonautonomous loaded linear ordinary differential equations with nonseparated multipoint and integral conditions. The method is based on the convolution of integral conditions into local ones. As a result, the original problem is reduced to an initial value (Cauchy) problem for systems of ordinary differential equations and linear algebraic equations. The approach proposed is used in combination with the linearization method to solve systems of loaded nonlinear ordinary differential equations with nonlocal conditions. An example of a loaded parabolic equation with nonlocal initial and boundary conditions is used to show that the approach can be applied to partial differential equations. Numerous numerical experiments on test problems were performed with the use of the numerical formulas and schemes proposed.     

Mean periodic solutions of a linear inhomogeneous first-order differential equation with random coefficients

We obtain deterministic first-order linear differential equations with ordinary and variational derivatives and deterministic initial conditions for the expectation and the second moment function of the solution of an ordinary scalar first-order linear inhomogeneous differential equation whose coefficients are random processes. We derive existence conditions for mean periodic solutions. In particular, we consider Gaussian and uniformly distributed random coefficients.     

Oscillation criteria for a forced mixed type Emden-Fowler equation with impulses

Some oscillation criteria for a forced mixed type Emden-Fowler equation with impulses are given. When the impulses are dropped, our results extend those of Sun and Meng [Y.G. Sun, F.W. Meng, Interval criteria for oscillation of second-order differential equations with mixed nonlinearities, Appl. Math. Comput. 15 (2008) 375-381], Sun and Wong [Y.G. Sun, J.S.W. Wong, Oscillation criteria for second order forced ordinary differential equations with mixed nonlinearities, J. Math. Anal. Appl. 334 (2007) 549-560] for second-order forced ordinary differential equation with mixed nonlinearities, Nasr [A.H. Nasr, Sufficient conditions for the oscillation of forced superlinear second order differential equations with oscillatory potential, Proc. Am. Math. Soc. 126 (1998) 123-125], Yang [Q. Yang, Interval oscillation criteria for a forced second order nonlinear ordinary differential equations with oscillatory potential, Appl. Math. Comput. 135 (2003) 49-64] for forced superlinear Emden-Fowler equation, Kong [Q. Kong, Interval criteria for oscillation of second-order linear differential equations, J. Math. Anal. Appl. 229 (1999) 483-492] for unforced second order linear differential equations, and Wong [J.S.W. Wong, Oscillation criteria for a forced second order linear differential equation, J. Math. Anal. Appl. 231 (1999) 235-240] for forced second order linear differential equation.     

Solving differential equations using modified Picard iteration

Many classes of differential equation are shown to be open to solution through a method involving a combination of a direct integration approach with suitably modified Picard iterative procedures. The classes of differential equations considered include typical initial value, boundary value and eigenvalue problems arising in physics and engineering and include non-linear as well as linear differential equations. Examples involving partial as well as ordinary differential equations are presented. The method is easy to implement on a computer and the solutions so obtained are essentially power series. With its conceptual clarity (differential equations are integrated directly), its uniform methodology (the overall approach is the same in all cases) and its straightforward computer implementation (the integration and iteration procedures require only standard commercial software), the modified Picard methods offer obvious benefits for the teaching of differential equations as well as presenting a basic but flexible tool-kit for the solution process itself.     

Lie symmetries and differential Galois groups of linear equations

For a linear ordinary differential equation the Lie algebra of its infinitesimal Lie symmetries is compared with its differential Galois group. For this purpose an algebraic formulation of Lie symmetries is developed. It turns out that there is no direct relation between the two above objects. In connection with this a new algorithm for computing the Lie symmetries of a linear ordinary differential equation is presented.     

A sufficient condition for GPN-stability for delay differential equations

Summary In this paper the author considers a linear test delay differential equation with non-constant coefficients related to the definitions of PN and GPN-stability for numerical methods. He defines stability properties for an ordinary differential equation together with stability properties of interpolants for numerical methods and in this way he gives sufficient conditions for GPN-stability.This work was supported by the Italian M.P.I. (funds 40%) and by C.N.R.