核实数据下非线性半参数EV模型的经验似然推断

考虑带有协变量误差的非线性半参数模型,借助于核实数据,本文构造了未知参数的三种经验对数似然比统计量,证明了所提出的统计量具有渐近X2分布,此结果可以用来构造未知参数的置信域．另外,本文也构造了未知参数的最小二乘估计量,并证明了它的渐近性质．仅就置信域及其覆盖概率的大小方面,通过模拟研究比较了经验似然方法与最小二乘法的优劣．     

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．     

纵向数据下变系数部分非线性模型的经验似然推断

考虑了纵向数据下变系数部分非线性模型中参数分量置信域的构造.为避免纵向数据组内相关性给构造经验似然比带来的困难,提出了参数分量的分块经验对数似然比统计量,证明了所构造的分块经验似然比统计量渐近于卡方分布.数据模拟表明所提出的分块经验似然方法在置信域大小和覆盖率方面要优于正态逼近方法.     

Panel Count Data模型参数的经验似然推断

对Panel Count Data的处理越来越受到人们的关注,Sun与Wei^([1-2])基于简单的半参数模型,提出了Panel Count Data的回归分析,并且给出了参数的估计方程。本文则基于经验似然的思想,讨论了上述Panel Count Data模型参数的置信域构造问题,特别仅通过经验似然置信区域给出了参数估计的方差阵估计,证明了估计的1/n相合性。基于Sun与Wei所给的数据,给出了参数置信区域的具体构造过程和结果。通过作图比较可以看出经验似然置信域要优于依据渐近正态性所构造的置信域。我们还依据所作出的经验似然置信域对参数估计的方差矩阵进行了估计,与用传统渐近正态性得到的矩阵较为接近。     

删失数据下单指标模型的经验似然推断

考虑删失数据下单指标模型, 研究了模型中参数的经验似然推断, 证明了所提出的调整的经验对数似然比渐近于卡方分布, 由此构造相应兴趣参数的置信域. 进一步, 由于模型中参数向量的范数等于1,利用该约束条件来降低参数的维数, 从而增加置信域的精度.模拟研究比较了经验似然方法和正态逼近方法的有限样本性质,从置信域的面积和覆盖概率两方面进行了比较,模拟结果表明经验似然方法优于正态逼近方法.     

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.     

纵向数据下部分非线性模型的广义经验似然推断

本文研究了纵向数据下部分非线性模型中未知参数的置信域的构造.利用经验似然方法,构造了非线性函数中未知参数的广义对数经验似然比统计量,证明了其渐近于卡方分布.同时,得到了未知参数的最大经验似然估计,并证明了其渐近正态性.     

缺失数据下半参数回归模型的经验似然推断

针对响应变量缺失下的半参数回归模型,构造模型中未知参数的经验对数似然比统计量,证明了所提出的统计量具有渐近χ2分布,由此构造未知参数的置信域,并就置信域的覆盖概率及区间长度方面,通过模拟研究与最小二乘法进行优劣比较.     

非线性半参数EV模型的经验似然置信域

考虑非参数协变量带有测量误差的非线性半参数模型,构造了模型中未知参数的经验对数似然比统计量,在测量误差分布为普通光滑分布时,证明了所提出的统计量具有渐近χ2分布,由此结果可以用来构造未知参数的置信域.另外也构造了未知参数的最小二乘估计量,并证明了它的渐近性质.就置信域及其覆盖概率大小方面,通过模拟研究比较了经验似然方法与最小二乘法的优劣.     

核实数据下响应变量缺失的线性EV模型经验似然推断

考虑响应变量随机缺失而协变量带有误差的线性模型,借助于核实数据和借补方法,构造了回归系数的两种经验似然比,证明了所提出的估计的经验对数似然比渐近于一个自由度为1的独立χ2变量的加权和;而经调整后所得的调整经验对数似然比渐近于自由度为p的χ2分布,该结果可以用来构造未知参数的置信域.此外,我们也构造了响应均值的调整经验对数似然比统计量,并证明了所提出的统计量渐近于x2分布,可用此结果构造响应均值的置信域.通过模拟研究比较了置信域的精度及其平均区间长度.     

Computing regions of interest for geometric features in digital images

We present an uncertainty model for geometric transformations based on polygonal uncertainty regions and transformation polytopes. The main contribution of this paper is a systematic approach for the computation of regions of interest for features by using the uncertainty model for affine and projective transformations. The focus is on the solution of transformation problems for geometric primitives, especially lines, so that regions of interest can be computed for corresponding geometric features in distinct images.     

論矩陣的一種變換

<正> 在本文中,數域限定為複數域.我們要來研究如下的變換:(1)(它將方陣A變成方陣B),式中P表示任意正則陣,P表示P的元素的共軛救構成的陣.所有的變换(1)顯然成羣.這種變換現在姑稱之為種變換.如果二方陣A與B可由一個種變換變此成彼,我們就說,A與B是對相似的.     

Comparison of Methods for the Computation of Multivariate t Probabilities

This article compares methods for the numerical computation of multivariate t probabilities for hyper-rectangular integration regions. Methods based on acceptance-rejection, spherical-radial transformations, and separation-of-variables transformations are considered. Tests using randomly chosen problems show that the most efficient numerical methods use a transformation developed by Genz for multivariate normal probabilities. These methods allow moderately accurate multivariate t probabilities to be quickly computed for problems with as many as 20 variables. Methods for the noncentral multivariate t distribution are also described.     

Decomposition of transformations

The decomposition theorem for transformations without any additivity assumptions is proved. This result is new even for image transformations. We also introduce a new class of transformations with some additivity assumpt ions which includes image transformations. Using the transformation from this class we generalize the Aarnes factorization theorem to representable deficient topological measures. We also establish the relationship between the decomposition of a representable deficient topological measure and the decomposition of the transformation mentioned above.     

Generalized Darboux Transformation for the Discrete Kadomtsev–Petviashvili Equation with Self-Consistent Sources

We construct several types of Darboux transformations for the discrete Kadomtsev–Petviashvili equation with self-consistent sources (dKPwS) including the elementary Darboux transformation, the adjoint Darboux transformation, and the binary Darboux transformation. These Darboux transformations can be used to obtain some solutions of the dKPwS. We give some solutions explicitly.     

Rectangle exchange transformations

We define rectangle exchange transformations analogously to interval exchange transformations. An interval exchange transformation is a mapping of the unit interval onto itself obtained by cutting the interval up into a finite number of subintervals and rearranging the pieces. A rectangle exchange transformation is a mapping of the unit square onto itself obtained by cutting the square up into a finite number of rectangular pieces and rearranging the pieces. We give a minimality condition for rectangle exchange transformations. We deal with various examples of ergodic rectangle exchange transformations. Related questions are discussed.With 2 Figures     

Bounded rank-1 transformations

We define the notion of canonical boundedness among rank-1 transformations and use it to characterize the class of bounded rank-1 transformations with trivial centralizer. We also explicitly characterize totally ergodic rank-1 transformations with bounded cutting parameter. Together with a recent result of Ryzhikov, our results provide a simple procedure for determining, purely in terms of the cutting and spacer parameters for the transformation, whether a bounded rank-1 transformation has minimal self-joinings of all orders.     

Quasi-measures with image transformations as generalized variables

We introduce an image transformation as a generalization of measurable maps. Structure properties of the image transformation is given. Unique extension of image transformations from solid sets is proven. The multidimensional median and sample median are presented as image transformations, providing a construction suitable for probability theorists.     

On the estimators of model-based and maximal reliability

Four estimators of the reliability for a composite score based on the factor analysis model and five estimators of the maximal reliability for the composite are presented. When the Wishart maximum likelihood is used for the estimation of the model parameters, it is shown that the five estimators of maximal reliability are the same. Asymptotic cumulants of the estimators and their logarithmic transformations are derived under arbitrary distributions with possible model misspecification. The theoretical results considering model misspecification when a model does not hold are shown to be closer to their simulated values than those neglecting model misspecification. Simulations of the confidence intervals using the normal approximation based on the asymptotically distribution-free theory and the asymptotic expansion by Hall’s method with variable transformation are performed.     

Cubic and Quartic Transformations of the Sixth Painlevé Equation in Terms of Riemann–Hilbert Correspondence

The starting point of this paper is a classification of quadratic polynomial transformations of the monodromy manifold for the 2 × 2 isomonodromic Fuchsian systems associated to the Painlevé VI equation. Up to birational automorphisms of the monodromy manifold, we find three transformations. Two of them are identified as the action of known quadratic or quartic transformations of the Painlevé VI equation. The third transformation of the monodromy manifold gives a new transformation of degree 3 of Picard’s solutions of Painlevé VI.