基于分层贝叶斯广义线性模型的小域估计方法研究

当前,构建恰当的小域估计方法是解决我国政府抽样调查中多层次推断问题的关键所在。由于小域的小样本特性,基于频率统计学的小域估计方法推断效果并不理想,而传统基于贝叶斯统计视角的小域估计方法在非连续型变量估计时适应性不强。本文在系统介绍传统贝叶斯小域估计方法的基站上,为了解决离散变量的估计推断问题,将广义线性模型引入到分层贝叶斯方法中,构建了基本的理论机制和分类数据的估计模型。基于此模型,运用全国流动人口动态监测调查2014年广东省内的样本数据进行实例测算,估计出广东省各地级市的流动人口学历分布情况,并将分层贝叶斯广义线性模型的估计结果与传统估计方法进行了对比分析。结果显示,分层贝叶斯广义线性模型在样本量充足的情况下能够准确地估计出目标小域的总体参数,在样本量不足的小域中依然能够给出稳健的估计结果。文章所构建的估计模型不仅可以充分利用先验信息和辅助信息,还适用于对复杂数据进行估计推断,能够为我国政府抽样调查的小域估计实践提供有价值的理论参考。     

基于混合地理加权Fay-Herriot模型的小域估计

作为一类区域层次模型, Fay-Herriot模型在小域估计中已经得到广泛的应用,这类模型假定各区域的直接估计是空间不相关的.很多情况下这个假定是不成立的,因此一些考虑空间效应的Fay-Herriot模型被提出.本文基于混合地理加权回归模型提出一类新的Fay-Herriot模型用以刻画空间非平稳性,基于提出的模型,给出小域目标参数的经验最佳线性无偏预测估计量,并研究了该估计量的均方误差.最后通过数值模拟验证了所提方法的有效性.     

测量误差模型的自适应LASSO变量选择方法研究

本文研究测量误差模型的自适应LASSO(least absolute shrinkage and selection operator)变量选择和系数估计问题.首先分别给出协变量有测量误差时的线性模型和部分线性模型自适应LASSO参数估计量,在一些正则条件下研究估计量的渐近性质,并且证明选择合适的调整参数,自适应LASSO参数估计量具有oracle性质.其次讨论估计的实现算法及惩罚参数和光滑参数的选择问题.最后通过模拟和一个实际数据分析研究了自适应LASSO变量选择方法的表现,结果表明,变量选择和参数估计效果良好.     

因变量缺失下变系数部分线性测量误差模型的变量选择

主要研究因变量存在缺失且协变量部分包含测量误差情形下,如何对变系数部分线性模型同时进行参数估计和变量选择.我们利用插补方法来处理缺失数据,并结合修正的profile最小二乘估计和SCAD惩罚对参数进行估计和变量选择.并且证明所得的估计具有渐近正态性和Oracle性质.通过数值模拟进一步研究所得估计的有限样本性质.     

基于混合地理加权回归模型的小域估计

近年来,基于空间模型的小域估计方法在抽样调查领域得到越来越多的关注.提出了一类混合地理加权单元层次小域模型用以刻画空间非平稳性,其中回归系数一部分为常数,另一部分则是随着地理位置的变化而变化.基于模型给出了域均值的估计,并研究了该估计量的均方误差及其估计.     

协变量区间删失时线性模型的参数估计

本文研究了当协变量为区间数据时的线性模型，通过构造区间数据变量的条件均值，得到了回归参数的估计，当协变量的分布已知时，证明了估计的无偏性与强相合性．时协变量的分布未知的情形也作了讨论．文中还作了若干模拟计算，从模拟的结果不难发现，利用本文提出的方法所获得的估计简便且具有较高的精度．     

一般线性混合模型中的最佳线性无偏估计和谱分解估计

该文在一般线性混合模型中, 研究了固定和随机效应线性组合的估计问题.对观测向量的协方差阵可以为奇异矩阵情形下,导出了该组合的最佳线性无偏估计,并证明了它的唯一性.在一般线性混合模型的特例, 三个小域模型下, 得到了小域均值u_i 和方差分量的谱分解估计. 进而, 获得了基于谱分解估计的两步估计均方误差的二阶逼近.     

响应变量缺失下非线性回归模型的加权半参数估计

基于逆概率加权方法研究了响应变量缺失下非线性回归模型的参数估计问题,提出了一种利用广义部分线性单指标模型对选择概率建模的加权半参数估计方法.从理论上证明了所得估计量具有渐近正态性,并通过数据模拟分析研究了所提方法在有限样本下的表现.     

带有不可忽略缺失数据的广义部分线性模型的贝叶斯分析

广义部分线性模型是广义线性模型和部分线性模型的推广,是一种应用广泛的半参数模型.本文讨论的是该模型在线性协变量和响应变量均存在非随机缺失数据情形下参数的Bayes估计和基于Bayes因子的模型选择问题,在分析过程中,采用了惩罚样条来估计模型中的非参数成分,并建立了Bayes层次模型;为了解决Gibbs抽样过程中因参数高度相关带来的混合性差以及因维数增加导致出现不稳定性的问题,引入了潜变量做为添加数据并应用了压缩Gibbs抽样方法,改进了收敛性;同时,为了避免计算多重积分,利用了M-H算法估计边缘密度函数后计算Bayes因子,为模型的选择比较提供了一种准则.最后,通过模拟和实例验证了所给方法的有效性.     

响应变量缺失下线性EV模型中参数的经验似然推断

考虑了响应变量随机缺失情形下的线性EV模型,通过利用逆概率加权的方法构造未知参数的经验对数似然比统计量,证明了所构造的经验对数似然比统计量渐近于X~2分布,利用这个结果可以构造未知参数的置信域     

Vehicle Scheduling — A Case Study

An application of vehicle scheduling techniques to the planning and operation of medical specimen collection services is described and a decision-making procedure suggested for use in this problem area. Some interesting results concerning the behaviour of vehicle scheduling savings algorithms, obtained during the course of the study, are also presented.     

Pushing Disks Together—The Continuous-Motion Case

If disks are moved so that each center—center distance does not increase, must the area of their union also be nonincreasing? We show that the answer is yes, assuming that there is a continuous motion such that each center—center distance is a nonincreasing function of time. This generalizes a previous result on unit disks. Our proof relies on a recent construction of Edelsbrunner and on new isoperimetric inequalities of independent interest. We go on to show analogous results for the intersection and for holes between disks. Received November 6, 1996, and in revised form June 16, 1997, and September 23, 1997.     

Mainframe or Micro? — A Case Study

This paper discusses the use of a computerised model for the allocation of a contract for operational work. The contract covers a defined period—normally two or three years, applies to one area or one Region of British Gas and is shared amongst many contractors. Millions of pounds are spent on all these contracts each year. The model was transferred from a mainframe to a microcomputer. The factors influencing this decision are presented, examined and seen to have aspects which apply generally. The study is used as an illustration of the problem of choosing between a mainframe and a microcomputer for an application.     

Staffing Rosters with Breaks — A Case Study

A complex manual rostering system for directory assistance, telephone operators has been rationalized and the technique of integer programming used to automatically generate staff rosters. These rosters can be generated by relatively unskilled clerical staff. They result in operator costs comparable to manually generated rosters and can be produced in a fraction of the time taken to produce their manual counterparts. Implementation is reported.     

Corporate Financial Planning in Insurance—A Case Study

ECGD's OR group has built a financial model to forecast accounts and performance indicators. This meets the need for soundly based financial planning, and its output is considered at the highest levels of the organization. The model enables senior management to evaluate the consequences of different courses of action against the background of varying world economic scenarios. Using a financial planning package has made for quick and efficient development of the model, which has been modified continually over the last 10 years. A flexible, adaptive approach has maintained the relevance of OR's contribution. The model forecasts the financial variables which underpin ECGD's quarterly financial review and annual business plan. The Department's Financial Policy and Planning Division uses the model extensively in its preparation of these documents and to explore feasible policies through ‘what-if’ analyses. Organizational changes, increasing delegation, and the introduction of strategic planning will require further adaptation of the model allowing access to a wider range of end-users.     

Adaptive Wavelet Methods II—Beyond the Elliptic Case

This paper is concerned with the design and analysis of adaptive wavelet methods for systems of operator equations. Its main accomplishment is to extend the range of applicability of the adaptive wavelet-based method developed in [17] for symmetric positive definite problems to indefinite or unsymmetric systems of operator equations. This is accomplished by first introducing techniques (such as the least squares formulation developed in [26]) that transform the original (continuous) problem into an equivalent infinite system of equations which is now well-posed in the Euclidean metric. It is then shown how to utilize adaptive techniques to solve the resulting infinite system of equations. This second step requires a significant modification of the ideas from [17]. The main departure from [17] is to develop an iterative scheme that directly applies to the infinite-dimensional problem rather than finite subproblems derived from the infinite problem. This rests on an adaptive application of the infinite-dimensional operator to finite vectors representing elements from finite-dimensional trial spaces. It is shown that for a wide range of problems, this new adaptive method performs with asymptotically optimal complexity, i.e., it recovers an approximate solution with desired accuracy at a computational expense that stays proportional to the number of terms in a corresponding wavelet-best N -term approximation. An important advantage of this adaptive approach is that it automatically stabilizes the numerical procedure so that, for instance, compatibility constraints on the choice of trial spaces, like the LBB condition, no longer arise.     

Scheduling of Telephone Betting Operators — A Case Study

A case study of the application of two O.R. techniques for determining optimal shift sizes for telephone betting operators is described. The first formulation uses linear programming and the second, network flow techniques. The network flow model has a considerable computational speed advantage, but the L.P. model is currently in operation.     

Scheduling the Cricket World Cup—a Case Study

This paper describes the scheduling of the matches for the 1992 World Cup of Cricket using a heuristic method programmed in Lotus 1-2-3. The many constraints of different types precluded an analytical approach and the authors describe their method and solution.     

Stock Control of Manufacturing Parts — A Case Study

The approach adopted for stock control of manufacturing parts in a small company is described. In particular, material requirements planning is compared with the standard stock control method previously in use, and safety buffering is considered. By altering the company's stock-holding and ordering policies a significant saving is achieved.     

Computing in a Medium-sized Company — A Case Study

An outline is given of the main tasks of a computer installed with the objective of improving managerial control of the production aspects of the Company. The function of OR relative to the computer is discussed both in determining the logic to be employed and the degree of "optimization" appropriate.