Rate of strong consistency of the maximum quasi-likelihood estimator in quasi-likelihood nonlinear models

Quasi-likelihood nonlinear models （QLNM） include generalized linear models as a special case. Under some regularity conditions, the rate of the strong consistency of the maximum quasi-likelihood estimation （MQLE） is obtained in QLNM. In an important case, this rate is O（n-^1/2（loglogn）^1/2）, which is just the rate of LIL of partial sums for i.i.d variables, and thus cannot be improved anymore.     

ALTERNATING DIRECTION FINITE ELEMENT METHOD FOR SOME REACTION DIFFUSION MODELS

This paper is concerned with some nonlinear reaction - diffusion models. To solve this kind of models, the modified Laplace finite element scheme and the alternating direction finite element scheme are established for the system of patrical differential equations. Besides, the finite difference method is utilized for the ordinary differential equation in the models. Moreover, by the theory and technique of prior estimates for the differential equations, the convergence analyses and the optimal L^2- norm error estimates are demonstrated.     

Consistency and asymptotic normality of the maximum quasi-likelihood estimator in quasi-likelihood nonlinear models with random regressors

This paper proposes some regularity conditions, which result in the existence, strong consistency and asymptotic normality of maximum quasi-likelihood estimator （MQLE） in quasi-likelihood nonlinear models （QLNM） with random regressors. The asymptotic results of generalized linear models （GLM） with random regressors are generalized to QLNM with random regressors.     

Adaptive quasi-likelihood estimate in generalized linear models

This paper gives a thorough theoretical treatment on the adaptive quasi-likelihood estimate of the parameters in the generalized linear models. The unknown covariance matrix of the response variable is estimated by the sample. It is shown that the adaptive estimator defined in this paper is asymptotically most efficient in the sense that it is asymptotic normal, and the covariance matrix of the limit distribution coincides with the one for the quasi-likelihood estimator for the case that the covariance matrix of the response variable is completely known.     

CONFIDENCE REGIONS IN TERMS OF STATISTICAL CURVATURE FOR AR（q） NONLINEAR REGRESSION MODELS

This paper constructs a set of confidence regions of parameters in terms of statistical curvatures for AR(q) nonlinear regression models. The geometric frameworks are proposed for the model. Then several confidence regions for parameters and parameter subsets in terms of statistical curvatures are given based on the likelihood ratio statistics and score statistics. Several previous results,, such as [1] and [2] are extended to AR(q)nonlinear regression models.     

SOME ASYMPTOTIC INFERENCE IN QUASI-LIKELIHOOD NONLINEAR MODELS:A GEOMETRIC APPROACH

§ 1　IntroductionIt is well known that quasi-likelihood models introduced by Wedderburn[1 ] greatlywiden the scope of generalized linear models by using a much weaker assumption in whichonly the firstand second moments ofresponse vector Yare needed to replace the full distri-butional assumption about Y in the models.It has drawn considerable attention in recentliterature(e.g.see[2～ 6] and so on) .However,little work has been done on the issuefrom a geometric viewpoint.The purpose of this p…     

Solitary Wave Solutions of Some Nonlinear Physical Models Using Riccati Equation Approach

Riccati equation approach is used to look for exact travelling wave solutions of some nonlinear physical models.Solitary wave solutions are established for the modified KdV equation,the Boussinesq equation and the Zakharov-Kuznetsov equation.New generalized solitary wave solutions with some free parameters are derived.The obtained solutions,which includes some previously known solitary wave solutions and some new ones,are expressed by a composition of Riccati differential equation solutions followed by a polynomial.The employed approach,which is straightforward and concise,is expected to be further employed in obtaining new solitary wave solutions for nonlinear physical problems.     

ITERATIVE QUASI-LIKELIHOOD FOR SEEMINGLY UNRELATED REGRESSION SYSTEMS

In the seemingly unrelated regression systems, the existing quasi-likelihood is always involved in the difficult problem of calculating inverse of a high order matrix specially for large systems. To avoid this problem, the new quasi-likelihood proposed in this paper is based mainly on a linearly iterative process of some unbiased estimating functions. Some finite sample properties and asymptotic behaviours for this new quasi-likelihood are investigated. These results show that the new quasi-likelihood for parameter of interest is E-sufficient, iteratively efficient and approximately efficient. Some examples are given to illustrate the theoretical results.     

TESTING FOR VARYING DISPERSION OF LONGITUDINAL BINOMIAL DATA IN NONLINEAR LOGISTIC MODELS WITH RANDOM EFFECTS

In this paper, it is discussed that two tests for varying dispersion of binomial data in the framework of nonlinear logistic models with random effects, which are widely used in analyzing longitudinal binomial data. One is the individual test and power calculation for varying dispersion through testing the randomness of cluster effects, which is extensions of Dean(1992) and Commenges et al (1994). The second test is the composite test for varying dispersion through simultaneously testing the randomness of cluster effects and the equality of random-effect means. The score test statistics are constructed and expressed in simple, easy to use, matrix formulas. The authors illustrate their test methods using the insecticide data (Giltinan, Capizzi & Malani (1988)).     

Geometry of Multinomial Distribution Models

In this paper, we present a new differential geometric framework to study mul-tinomial distribution models in which a p-vector parameter θ is of interest. Thegeometry and models discussed here are different from those of Efron and Amari(EA) and Bates and Watts (BW). In brief, the models we study are close to thoseof EA and the geometry we present is close to that of BW. We try to combine bothEA and BW advantages into our models. Suppose that Y = (y_1,…,y_n)~T are multinomial random variables whose densityfunction has the form     

SOME ASYMPTOTIC INFERENCE IN MULTINOMIAL NONLINEAR MODELS （A GEOMETRIC APPROACH）

ＳＯＭＥＡＳＹＭＰＴＯＴＩＣＩＮＦＥＲＥＮＣＥＩＮＭＵＬＴＩＮＯＭＩＡＬＮＯＮＬＩＮＥＡＲＭＯＤＥＬＳ（ＡＧＥＯＭＥＲＩＣＡＰＰＲＯＡＣＨ）￥ＷＥＩＢＯＣＨＥＮＧ（ＤｅｐａｒｔｍｅｎｔｏｆＭａｔｈｅｍａｔｉｃｓ，ＳｏｕｔｈｅａｓｔＵｎｉｖｅｒｓｉｔｙ...     

SOME ASYMPTOTIC INFERENCE ON NONLINEAR MODELS WITH RANDOM EFFECTS （A GEOMETRIC APPROACH）

ＳＯＭＥＡＳＹＭＰＴＯＴＩＣＩＮＦＥＲＥＮＣＥＯＮＮＯＮＬＩＮＥＡＲＭＯＤＥＬＳＷＩＴＨＲＡＮＤＯＭＥＦＦＥＣＴＳ（ＡＧＥＯＭＥＴＲＩＣＡＰＰＲＯＡＣＨ）ＺＨＯＮＧＸＵＰＩＮＧＡＮＤＷＥＩＢＯＣＨＥＮＧＡｂｓｔｒａｃｔ．Ａｇｅｏｍｅｔｒｉｃｆｒａｍｅ...     

非线性再生散度随机效应模型参数置信域的曲率表示

该文基于Laplace逼近建立了非线性再生散度随机效应模型在Euclid空间中的几何结构, 并在此基础上研究了此模型参数和子集参数的置信域, 进一步推广和发展了 Hamilton, Watts 和 Bates^[1]关于正态非线性回归模型, Wei^[2,3]关于嵌入模型和指数族非线性模型, Zhu, Tang 和 Wei^[4]关于半参数非线性模型,唐年胜、韦博成和王学仁^[5]关于非线性再生散度模型, Tang 和 Wang^[6]关于拟似然非线性模型等的结果.     

半参数非线性回归模型渐近推断的几何

本文利用Severini和Wong^[1]提出的最佳偏差曲线的概念,对半参数非线性回归模型建立了类似于Bates和Watts^[2]的几何结构。利用这个几何结构,我们研究了与统计曲率有关的某些渐近推断。文献中的许多结果^[3-6]被推广到半参数非线性回归模型。     

误差非线性回归模型基于几何方法的若干二阶渐近性质

该文用微分几何方法对AR(q)误差非线性回归模型若干二 阶渐近性质进行了研究. 作者基于Fisher信息阵在欧氏空间定义了内积，并在期望参数空间建立了几何结构. 基于上述几何结构，给出了AR(q)误差非线性回归模型若干二阶渐近性质的曲率表示. 将前人的一些结果推广到AR(q)误差非线性回归模型.      

非线性随机效应模型的置信域

本文对非线性随机效应模型,建立了微分几何框架,推广了Bates＆Wates关于非线性模型几何结构．在吡基础上,我们导出了关于固定效应参数和子集参数的置信域的曲率表示,这些结果是BatesandWates(1980),Hamilton(1986)与Wei(1994)等的推广．     

非线性再生散度模型基于统计曲率的诊断统计量

对非线性散度模型在Euclid空间建立几何结构。在此基础上，研究了均值漂移模型的曲率度量。从而导出相应Cook距离，似然距离等诊断统计量的二阶近似公式。     

带寿命数据非线性随机效应模型的置信域(英文)

本文对带寿命数据非线性随机效应模型,建立了微分几何框架,推广了Bates Wates关于非线性模型几何结构.在此基础上,我们导出了关于固定效应参数和子集参数的置信域的曲率表示,这些结果是Bates and Wates(1980),Hamilton(1986)和Wei(1998)等的推广.