基于最佳线性无偏估计的模型权数的小域估计

文章介绍了一种基于EBLUP的模型权数的小域估计方法.这种估计方法一方面使小域的目标估计量是加权线性组合,从而使估计过程以及均方误的估计更加简单;另一方面得到的估计量不依赖于模型的假定,是一种稳健的估计量.文章还通过一个简单的模拟案例说明了这种估计量的稳健的性质,说明这种估计方法是一种非常符合实际调查情况的小域估计方法.     

放回抽样下HT估计量的性质及应用

放回抽样下传统的估计方法是采用Hansen-Hurwitz估计。而放回抽样下HH估计量并不是一致最小方差无偏估计,本文提出了另一种估计方法,即采用Horvitz-Thompson估计,并论证了放回抽样下HT估计量的三条定理,及与HH估计量的比较。然后以放回简单随机抽样和PPS抽样为例,通过理论公式、计算机模拟以及具体案例,进行更具体的分析。说明在一定条件下,HT估计量相对更优。在实际应用中,本文也提出了通过比较方差估计作为选取估计量的准则。     

基于多重抽样框的校准估计方法研究

在现代抽样调查中,校准估计方法能够通过有效利用辅助信息来提高估计量的精度,多重抽样框抽样调查则不仅可以解决单一抽样框覆盖不全的问题,还可以节约抽样设计阶段的成本。本文将这两种现代抽样估计与设计方法进行结合,将校准估计方法引入到基于多重抽样框的抽样调查体系中,在实现节约调查成本的同时,还能够提高估计量的精度。文章首先按照分离抽样框与组合抽样框估计方法的分类思路,对传统多重抽样框估计方法进行系统梳理;然后在最短距离法校准估计的分析框架下,按照调查时所能掌握辅助信息的具体情况,给出了两类多重抽样框估计情形下的各种不同形式的校准估计量;随后数值分析的比较结果也表明在多重抽样框中校准估计量的估计效率明显优于传统估计量;最后对本文研究进行总结的基础上,给出了我国抽样实践中应用这套先进抽样估计方法体系的展望。     

抽样调查中基于模型的稳健预测方法

基于模型的推断是抽样技术中推断估计量的一种重要方式。文章研究得出,当比率估计模型或者扩张估计模型偏离总体真实模型时,比率估计和扩张估计往往是有偏的,平衡样本能够消除比率估计和扩张估计的偏倚,使得估计量是偏倚稳健的。     

连续性抽样调查中总体均值的估计

在连续性抽样调查中,利用前期信息和辅助信息可以大大提高估计精度,但是以往的估计量大多假设辅助信息总体均值已知,文章介绍一种在连续性抽样调查中,辅助信息总体均值未知的情况下,通过两阶段抽样,利用轮换样本方法和辅助变量信息,对总体均值进行估计的新的估计方法,并将新提出的估计量与原有的估计量进行比较,发现其精度更高,而且有利于减少调查成本。     

四分之一轮换面板下的稳健估计方法

文章从过度置信度视角考察了广义最小二乘估计量在四分之一轮换面板下产生的偏误问题,并提出了一种稳健估计方法来修正过高的过度置信度,进而提高估计精度。在一定的设计条件下,证明了修正后的估计量具有一致性和渐近正态分布特征等优良性质。模拟研究结果显示,与四分之一轮换面板下广义最小二乘估计量相比,提出的估计方法在保持相对偏差和均方误差基本不变的情况下,有效降低了过度置信度。     

基于GMM的缺失数据回归模型的半参数估计

响应变量存在数据缺失的情况广泛出现在社会经济研究中,对响应变量存在数据缺失的回归模型提出了一个在矩估计框架下的单一的半参数估计量,这种估计量保留了参数回归估计量与非参数匹配估计量的特性,从而使得该估计量既能在响应变量被观测的子样本中保持较好的拟合性,又能够降低响应变量未被观测的子样本的估计误差,并且证明了这种估计量是一致、渐进正态估计量。     

使用变异系数和Kurtosis系数的双辅助变量的比估计

文章以提高估计量的精度为目的,定义了一种新的使用两个辅助变量的比估计法,研究了这种方法下估计量的均方误差,证明了在一定条件下这种方法下的估计量优于传统的使用双辅助信息的比估计量,同时对这些估计量进行了比较。并且从数值角度得到了验证。     

时变系数空间自回归面板数据模型的极大似然估计

本文对扰动项存在跨时期的异方差、但不存在序列相关的时变系数空间自回归模型提出了极大似然的估计方法,并证明了该估计量的一致性,同时,证明了该估计量渐进服从正态分布,由此说明该估计量具有优良的大样本性质。同时,我们还对本文所提出估计量的小样本性质进行了数值模拟。本文研究表明,估计量虽然在N较小时偏差较大,但是随着N的不断增加,估计量偏差减小,体现了比较优良的渐进性质。同时,估计量的偏差会随着时期数的增加而变大,这说明本文所提出的估计方法适用于个体数较多、时期数较少的短面板数据。     

基于γ散度的单元水平模型小域稳健估计

在基于抽样调查数据对总体参数进行估计的方法中，小域估计方法能够借助于辅助信息对小样本乃至无样本区域的参数进行有效的估计，并被广泛应用于抽样估计领域。单元水平模型作为小域估计的基本模型之一，是处理单元级别数据估计的有力工具之一。在单元水平模型的应用条件中，需假定区域随机误差和模型随机误差均服从正态分布。然而，在抽样调查中，满足这一条件的调查数据是很少的，尤其是在观测数据中出现离群值时。不满足正态性假设条件下的小域估计量会产生较大的偏差和均方误，因此有必要研究针对正态性假设和离群观测值不敏感的稳健估计方法。通过引入γ散度和γ似然函数，构建了基于单元水平模型的小域稳健估计方法，得到了模型参数的稳健估计和小域目标变量的稳健估计。与现有的稳健估计方法相比，所提新方法能更好地处理区域随机误差和模型随机误差非正态的情形，对于目标变量存在离群观测的情形，具有更好的稳健性，估计均方误更小。在利用模拟数据进行验证中，比较了不同误差分布情形下几类常用估计方法得到的估计量的均方误差，并进一步探究了随着污染分布的方差和比率变化，所得估计量的均方误差变化情形。最后，通过应用于经典的小域估计数据，进一步验证了所提新...     

随机效应空间滞后单指数面板模型

对随机效应空间滞后单指数面板模型,本文构建了该模型的截面极大似然估计方法,从理论证明和数值模拟两方面分别考察了其估计量的大样本性质和小样本表现。研究结果表明:(1)在大样本条件下,估计量均具有一致性,并且参数估计量具有渐近正态性。(2)在小样本条件下,各估计量依然具有良好的表现,其精度随着样本容量的增加而提高;空间权重矩阵结构的复杂性对空间相关系数的估计量影响较大,但对其他估计量的影响较小。     

Bayes and admissibility properties of estimators in truncated parameter spaces

For the problem of estimating a parameter θ when θ is known to lie in a closed, convex subset D of R^k, conditions are given under which estimators δ of θ cannot be Bayes estimators, as well as conditions under which δ is inadmissible. The estimators considered are so-called “boundary estimators”. Maximum-likelihood estimators in truncated parameter spaces are examples to which our results often apply. For the special case when k = 1 and D is compact, two classes of estimators dominating the inadmissible ones are constructed. Some examples are given.     

A new geometric first-order integer-valued autoregressive (NGINAR(1)) process

A new stationary first-order integer-valued autoregressive process with geometric marginal distributions is introduced based on negative binomial thinning. Some properties of the process are established. Estimators of the parameters of the process are obtained using the methods of conditional least squares, Yule–Walker and maximum likelihood. Also, the asymptotic properties of the estimators are derived involving their distributions. Some numerical results of the estimators are presented with a discussion to the obtained results. Real data are used and a possible application is discussed.     

Bayesian estimation of mean and square of mean of normal distribution using linex loss function

In this paper, the Bayes estimators for mean and square of mean ol a normal distribution with mean μ and vaiiance σ r2 (known), relative to LINEX loss function are obtained Comparisons in terms of risk functions and Bayes risks of those under LINEX loss and squared error loss functions with their respective alternative estimators viz, UMVUE and Bayes estimators relative to squared error loss function, are made. It is found that Bayes estimators relative to LINEX loss function dominate the alternative estimators m terms of risk function snd Bayes risk. It is also found that if t2 is unknown the Bayes estimators are still preferable over alternative estimators.     

Estimation in the singly truncated weibull distribution with an unknown truncation point

When truncation points are unknown, they must be treated as additional parameters to be estimated from the sample data. In this article, estimators are derived for the truncation parameter in addition to basic parameters of both 1eft and riqht sing1y truncated Weibull distributions, Maximum likelihood estimators and estimators involving expected values of appropriate order statistics are derived, Asymptotic sampling errors of estimates are also given, Ill ustrative examples are inc1uded.     

Fitting type I generalized Logistic distribution by modified method based on percentiles

ABSTRACT

In this paper, we propose a modified estimators based on percentiles (MPCE) to improve the estimators performance based on percentiles for the generalized Logistic distribution. Simulation results indicate that MPCE outperforms other existing methods in terms of MSE. Finally, the proposed method is applied to a real data set.     

Bias-adjustment and calibration of jackknife variance estimator in the presence of non-response

In this paper, bias-adjustment in the jackknife estimator of variance accredited to Rao and Sitter (1995) has been considered. Then the bias-adjusted Rao and Sitter (1995) estimator has been calibrated such that its expected value under the imputing superpopulation model remains the same as the expected value of the mean squared error of the ratio estimator in the presence of non-response. A simulation study has been performed to compare the six different estimators of variance: out of them four estimators belong to Rao and Sitter (1995) and the other two proposed estimators are named as bias-adjusted and bias-adjusted-cum-calibrated estimators. The empirical relative bias and empirical relative efficiency of the two proposed estimators with respect to the four existing estimators accredited to Rao and Sitter (1995) have been investigated through simulations. The bias-adjusted-cum-calibrated estimator has been found to be an efficient estimator in the case of heteroscadastic populations. The present paper considers the situation of simple random and without replacement sampling. The possibility of obtaining a negative estimate of variance by the estimator due to Kim et al. (2006) has been pointed out.     

Estimation of Mean of Sensitive Characteristics for Successive Sampling

In this article, new estimators for estimating the population mean of a sensitive variable using the concept of successive sampling over two occasions are proposed. The unbiasedness and the variance properties of the proposed estimators are investigated analytically as well as numerically.     

On the efficiency property of the linear estimators in the bivariate pearson type vii distribution

In this paper the efficiency property of the estimators of the parameters of the bivariate Pearson type VII distribution is studied inside the family of linear estimators, assuming that the sample is constituted by dependent random vectors. It is proven that, although there are not efficient linear estimators, the sample mean and the sample covariance matrix (affected by an unbiasedness weighting) are unbiased linear estimators of minimum distance to the Cramér-Rao lower bound. Finally, a numerical simulation example shows that the proposed estimators are computationally feasible.     

A necessary and sufficient condition for an estimator to be optimal

A necessary condition for an estimator δ₀ to be optimal in a class Δ of estimators is derived. Thic condition becomes also sufficient when the loss function is convex.