Assessing the performance of empirical bayes estimators

Methods for deriving empirical Bayes estimators are generally available. Corresponding general techniques for assessing the performance of these estimators are not widely developed yet, however. In this paper we provide a general procedure for assessing and comparing the performance of the empirical Bayes estimators and other estimators in a given data set.     

Simultaneous Estimation of Several Intraclass Correlation Coefficients

Based on shrinkage and preliminary test rules, various estimators are proposed for estimation of several intraclass correlation coefficients when independent samples are drawn from multivariate normal populations. It is demonstrated that the James-Stein type estimators are asymptotically superior to the usual estimators. Furthermore, it is also indicated through asymptotic results that none of the preliminary test and shrinkage estimators dominate each other, though they perform relatively well as compared to the classical estimator. The relative dominance picture of the estimators is presented. A Monte Carlo study is performed to appraise the properties of the proposed estimators for small samples.     

Asymptotic properties of some subset vector autoregressive process estimators

We establish consistency and derive asymptotic distributions for estimators of the coefficients of a subset vector autoregressive (SVAR) process. Using a martingale central limit theorem, we first derive the asymptotic distribution of the subset least squares (LS) estimators. Exploiting the similarity of closed form expressions for the LS and Yule–Walker (YW) estimators, we extend the asymptotics to the latter. Using the fact that the subset Yule–Walker and recently proposed Burg estimators satisfy closely related recursive algorithms, we then extend the asymptotic results to the Burg estimators. All estimators are shown to have the same limiting distribution.     

Safe and tight linear estimators for global optimization

Global optimization problems are often approached by branch and bound algorithms which use linear relaxations of the nonlinear constraints computed from the current variable bounds. This paper studies how to derive safe linear relaxations to account for numerical errors arising when computing the linear coefficients. It first proposes two classes of safe linear estimators for univariate functions. Class-1 estimators generalize previously suggested estimators from quadratic to arbitrary functions, while class-2 estimators are novel. When they apply, class-2 estimators are shown to be tighter theoretically (in a certain sense) and almost always tighter numerically. The paper then generalizes these results to multivariate functions. It shows how to derive estimators for multivariate functions by combining univariate estimators derived for each variable independently. Moreover, the combination of tight class-1 safe univariate estimators is shown to be a tight class-1 safe multivariate estimator. Finally, multivariate class-2 estimators are shown to be theoretically tighter (in a certain sense) than multivariate class-1 estimators.     

Second derivative estimation using harmonic analysis

Simulation sensitivity analysis is an important problem for simulation practitioners analyzing complex systems. The significance of this problem has resulted in the development of various gradient estimators that can be used to address this issue. Although higher derivative estimators have been discussed concurrently, less attention has been given to assess the efficiency and feasibility of computing such estimators. In this paper, two second derivative estimators are presented. The first estimators, called the HFD estimators, combine harmonic gradient estimators with finite differences second derivative estimators. The resulting hybrid estimators requireO(p) fewer simulation runs to implement compared to the straightforward finite differences approach, wherep is the number of input parameters in the simulation model. The second estimators, called the HA estimators, incorporate harmonic analysis directly, requiring one or two simulation runs to implement, depending on whether a control variate simulation run is made. Expressions for the bias and the variance of the HFD and the HA estimators (with and without variance reduction techniques) are derived. Optimal mean squared error convergence rates are also discussed. In particular, the convergence rates for both these estimators are shown to be the same, though the computational performance of the HFD estimators is better than that for the HA estimators on anM/M/1 queue simulation model. Computational results for the HFD estimators on an (s, S) inventory system simulation model are also included.     

Rates of convergence for minimum contrast estimators

Summary We shall present here a general study of minimum contrast estimators in a nonparametric setting (although our results are also valid in the classical parametric case) for independent observations. These estimators include many of the most popular estimators in various situations such as maximum likelihood estimators, least squares and other estimators of the regression function, estimators for mixture models or deconvolution... The main theorem relates the rate of convergence of those estimators to the entropy structure of the space of parameters. Optimal rates depending on entropy conditions are already known, at least for some of the models involved, and they agree with what we get for minimum contrast estimators as long as the entropy counts are not too large. But, under some circumstances (large entropies or changes in the entropy structure due to local perturbations), the resulting the rates are only suboptimal. Counterexamples are constructed which show that the phenomenon is real for non-parametric maximum likelihood or regression. This proves that, under purely metric assumptions, our theorem is optimal and that minimum contrast estimators happen to be suboptimal.     

Laws of the iterated logarithm for nonparametric density estimators

Summary We establish a law of the iterated logarithm for a triangular array of independent random variables, and apply it to obtain laws for a large class of nonparametric density estimators. We consider the case of Rosenblatt-Parzen kernel estimators, trigonometric series estimators and orthogonal polynomial estimators in detail, and point out that our technique has wider application.     

基于价格极值构建有效价差的广义矩估计

基于Corwin和Schultz(2012)提出的有效价差的High-Low估计,结合价格极值信息得到新的一阶矩条件,构造了有效价差的广义矩估计。随后通过随机数值模拟比较了基于价格极值的广义矩估计(GMM)与Roll的协方差估计、Bayes估计以及Corwin和Schultz的High-Low估计在多种不同状态下的估计精度。数值模拟结果显示,无论在交易连续的理想状态下还是交易不连续且波动率相对不高的非理想状态下,GMM估计的精度均高于其余三种估计;基于我国股票市场的实例分析,也表明GMM估计的估计精度优于其余三种估计。因此,GMM估计为度量金融资产的交易成本提供了一种有效方法。     

多元失效时间删失数据边际风险模型加权估计的渐近正态性

对于多元失效时间数据,可以根据工作独立的假定来估计边际风险模型中的未知参数,但工作独立方法通常会失去估计的效率.为了充分利用不同失效类型之间的潜在相关性,提高估计的效率,可以通过加权的方法给出参数的加权部分似然估计.然而由于多元失效数据是高维数的数据,选择最优权是困难的.因此,Fan,Zhou,Cai和Chen曾基于参数估计向量中每个元的方差提出了一些次优加权方法,然后从参数向量所有分量估计的角度出发,构造了未知参数的复合加权部分似然估计,但他们没有给出这些复合加权估计的渐近性质.本文将对复合加权部分似然估计进一步的研究,推导了这个估计的渐近正态性,并给出了该估计的协方差阵以及协方差估计.同时,将该方法应用于艾滋病临床试验的实际数据,给出了有意义的解释和说明.最后进行了相关估计的一些数值模拟计算.     

Linearized Regression Model with Constraints of Type II

A linearization of the nonlinear regression model causes a bias in estimators of model parameters. It can be eliminated, e.g., either by a proper choice of the point where the model is developed into the Taylor series or by quadratic corrections of linear estimators. The aim of the paper is to obtain formulae for biases and variances of estimators in linearized models and also for corrected estimators.     

All admissible linear estimators of a regression coefficient under a balanced loss function

Admissibility of linear estimators of a regression coefficient in linear models with and without the assumption that the underlying distribution is normal is discussed under a balanced loss function. In the non-normal case, a necessary and sufficient condition is given for linear estimators to be admissible in the space of homogeneous linear estimators. In the normal case, a sufficient condition is provided for restricted linear estimators to be admissible in the space of all estimators having finite risks under the balanced loss function. Furthermore, the sufficient condition is proved to be necessary in the normal case if additional conditions are assumed.     

Robustness of Stein-type estimators under a non-scalar error covariance structure

The Stein-rule (SR) and positive-part Stein-rule (PSR) estimators are two popular shrinkage techniques used in linear regression, yet very little is known about the robustness of these estimators to the disturbances’ deviation from the white noise assumption. Recent studies have shown that the OLS estimator is quite robust, but whether this is so for the SR and PSR estimators is less clear as these estimators also depend on the F statistic which is highly susceptible to covariance misspecification. This study attempts to evaluate the effects of misspecifying the disturbances as white noise on the SR and PSR estimators by a sensitivity analysis. Sensitivity statistics of the SR and PSR estimators are derived and their properties are analyzed. We find that the sensitivity statistics of these estimators exhibit very similar properties and both estimators are extremely robust to MA(1) disturbances and reasonably robust to AR(1) disturbances except for the cases of severe autocorrelation. The results are useful in light of the rising interest of the SR and PSR techniques in the applied literature.     

Classes of improved estimators for parameters of a Pareto distribution

The problem of estimating parameters of a Pareto distribution is investigated under a general scale invariant loss function when the scale parameter is restricted to the interval (0, 1]. We consider the estimation of shape parameter when the scale parameter is unknown. Techniques for improving equivariant estimators developed by Stein, Brewster–Zidek and Kubokawa are applied to derive improved estimators. In particular improved classes of estimators are obtained for the entropy loss and a symmetric loss. Risk functions of various estimators are compared numerically using simulations. It is also shown that the technique of Kubokawa produces improved estimators for estimating the scale parameter when the shape parameter is known.     

On optimal estimators of shift population density functions and conditions of their consistency

A new technique of optimal estimation of density functions for exponential shift families on a homogeneous space of a Lie group is proposed. In contrast to traditional methods, the approach considered is essentially based on the algebraic properties of shift families. Here we give a universal formula for consistent estimators of density functions covering different classes of estimators such as unbiased estimators with uniformly minimum variance and Bayesian estimators under two popular loss functions. The representations of some maximal invariant density functions are derived and simultaneously a close connection between the estimators and these density functions is established. Proceedings of the XVII Seminar on Stability Problems for Stochastic Models, Kazan, Russia, 1995, Part III.     

非零截距线性和非线性回归模型的T-型稳健估计算法

首先给出非零截距线性模型T-型估计的模型与EM算法,其次给出非线性回归模型参数的T-型估计,利用泰勒级数对模型线性化,得到参数估计的迭代算法,最后用数值模拟实验验证了该算法的正确性和证实了T-型估计的稳健性.     

Pitman准则下指数分布刻度参数幂的分组定数截尾估计

在Pitman准则下,对指数分布刻度参数σ的形为σ     

Asymptotically optimal estimation in the linear regression problem in the case of violation of some classical assumptions

We consider the problem of estimating the unknown parameters of linear regression in the case when the variances of observations depend on the unknown parameters of the model. A two-step method is suggested for constructing asymptotically linear estimators. Some general sufficient conditions for the asymptotic normality of the estimators are found, and an explicit form is established of the best asymptotically linear estimators. The behavior of the estimators is studied in detail in the case when the parameter of the regression model is one-dimensional.     

Estimation of parameters in the fractional compound Poisson process

This paper discusses method-of-moments estimators for parameters in the fractional compound Poisson process and establishes asymptotic normality of estimators. Simulation are presented to illustrate the properties of the estimators.     

Estimating means from a non-I.I.D. Mixture of Poisson samples

In this paper, a family of estimators for estimating means when mixing two independent Poisson samples is proposed. This family is based on the probability-generating function of the Poisson distribution and is offered as an alternative to the maximum likelihood estimators, which have some drawbacks. These estimators include the method of moments estimators as a special limiting case.     

Nonparametric estimation of the spectral measure,and associated dependence measures,for multivariate extreme values using a limiting conditional representation

The traditional approach to multivariate extreme values has been through the multivariate extreme value distribution G, characterised by its spectral measure H and associated Pickands’ dependence function A. More generally, for all asymptotically dependent variables, H determines the probability of all multivariate extreme events. When the variables are asymptotically dependent and under the assumption of unit Fréchet margins, several methods exist for the estimation of G, H and A which use variables with radial component exceeding some high threshold. For each of these characteristics, we propose new asymptotically consistent nonparametric estimators which arise from Heffernan and Tawn’s approach to multivariate extremes that conditions on variables with marginal values exceeding some high marginal threshold. The proposed estimators improve on existing estimators in three ways. First, under asymptotic dependence, they give self-consistent estimators of G, H and A; existing estimators are not self-consistent. Second, these existing estimators focus on the bivariate case, whereas our estimators extend easily to describe dependence in the multivariate case. Finally, for asymptotically independent cases, our estimators can model the level of asymptotic independence; whereas existing estimators for the spectral measure treat the variables as either being independent, or asymptotically dependent. For asymptotically dependent bivariate random variables, the new estimators are found to compare favourably with existing estimators, particularly for weak dependence. The method is illustrated with an application to finance data.