含非随机缺失数据的面板数据参数估计方法

抽样调查领域常采用对多个受访者进行跟踪调查得到面板数据,进而对总体特性进行统计推断,在面板数据中常含缺失数据,大多数处理面板缺失数据的软件都是直接删去含缺失值的受访者以得到完全数据集,当数据缺失机制为非随机缺失时会导致总体参数估计结果有偏。本文针对数据缺失机制为非随机缺失情形下,如何对面板数据进行统计分析进行了阐述,主要采用的是基于模型的似然推断法,对目标变量、缺失指示变量和随机效应向量的联合分布建模,在已有选择模型和模式混合模型的基础上,引入随机效应,研究目标变量期望的计算方法,并研究随机效应杂合模型下参数的估计方法,在变量分布相对简单的情形下给出了用极大似然法推断总体参数的估计步骤,最后通过模拟分析比较方法的优劣。     

多因素模型中的最佳投资比例

文章推导出多因素模型在允许卖空和不允许卖空的情况下的最佳投资比例,推广了单因素模型假设下EGP方法最佳投资比例的计算公式;并对上海股票市场一个投资组合进行实证分析。     

基于分层模型的缺失数据插补方法研究

大规模抽样调查多采用复杂抽样设计,得到具有分层嵌套结构的调查数据集,其中不可避免会遇到数据缺失问题,针对分层结构含缺失数据集的插补策略目前鲜有研究。本文将Gibbs算法应用到分层含缺失数据集的多重插补过程中,分别研究了固定效应模型插补法和随机效应模型插补法,进而通过理论推导和数值模拟,在不同组内相关系数、群组规模、数据缺失比例等情形下,从参数估计结果的无偏性和有效性两方面,比较不同方法的插补效果,给出插补模型的选择建议。研究结果表明,采用随机效应模型作为插补模型时,得到的参数估计结果更准确,而固定效应模型作为插补模型操作相对简便,在数据缺失比例较小、组内相关系数较大、群组规模较大等情形下,可以采用固定效应插补模型,否则建议采用随机效应插补模型。     

有限总体的估计——基于超总体模型

抽样调查下有限总体的估计，一般是基于传统抽样设计，另一种是基于超总体模型，即假定总体取值不是确定的，而是由超总体模型产生的。本文以简单随机抽样和抽样为例，揭示了在不同情况下，如何得到超总体模型下有限总体的估计，并对基于设计和基于模型两种观点进行了比较分析。     

权数对基于模型推断的影响分析

利用抽样调查数据对总体参数进行推断通常分为两种途径：一种是基于设计的推断体系；另一种是基于模型的推断体系。基于设计的推断以随机化理论为基础，推断依赖于抽样设计，在大样本下估计量具有无偏性和一致性，但在样本量较小或存在非抽样误差等情况下效率较低。基于模型的推断认为有限总体是一个来自无限超总体的随机样本，推断依赖于模型假设，构建超总体模型具有很大的灵活性，有利于充分利用总体辅助信息并提高估计精度，但在模型假定有误或样本的入样过程不具有无信息性时存在估计误差。如何将两种推断途径相结合，在体现样本对总体代表性的同时，保证估计效率和估计量的优良性质，尚待研究。权数在基于设计的推断中起着核心作用，能够反映抽样设计对样本的影响，实现样本对总体的还原。将权数引入基于模型的推断，可以使基于模型推断的结果具有总体代表性，能更好地发挥两种推断体系的组合优势，并削弱模型假定对推断效果的影响。据此，从权数对于模型推断的影响入手，针对因果推断问题，提出将权数同时引入倾向得分模型和预测模型的建模过程，来构造双稳健估计的方法，并通过模拟研究加以验证。最终结果表明，根据文章所提出的方法进行处理效应的估计，能够充分发挥权数的作用，得到更准确、更稳健的估计结果。实证部分采用2017年CGSS调查数据进行分析，进一步说明在基于调查数据进行模型推断时应充分考虑抽样设计的影响，为科研人员进行因果推断以及其他基于调查数据开展的研究提供参考。     

春节模型的设计与应用

 季节调整是经济数据预处理中非常重要的一个步骤。现有的主流季节调整方法X-12-ARIMA 和TRAMO/SEATS中都包含节假日因素的调整。由于不同的国家节假日一般不同,因此各国在进行经济数据的季节调整时,都需要结合本国的假日对季节调整方法进行修正。春节是中国最为重要而且持续时间最长的节日,具体日期可以出现在一月也可以在二月。本文基于X-12-ARIMA方法,同时考虑春节对经济指标的正负性影响效应、春节影响的变化速率以及春节效应的时长三个因素,设计了十二个不同类型的春节模型。本文应用Eviews软件和Demetra软件,采集不同的经济指标,对所设计的春节模型进行了应用研究,并根据异常值改善标准,对最佳的春节模型进行了选择与比较分析。     

FDI对我国各地区经济增长的非线性效应分析

本文运用面板平滑转换(PSTR)模型研究国外直接投资(FDI)对我国各地区经济增长的非线性效应。选取了FDI及影响FDI的市场规模、基础设施、产业结构等8个变量作为解释变量构建单因素和多因素PSTR模型。结果表明我国经济增长与FDI之间存在着平滑转换机制效应;FDI对经济增长的影响是非线性的;FDI是我国各地区经济增长的充分条件,但不是必要条件。     

中国经济增长中的人力资本门槛效应研究

Azariadis和Drazen(1990)的人力资本门槛外部性理论研究发现,引入人力资本后两部门增长模型存在多重稳态均衡,人力资本积累跨越某一门槛值后经济体进入新的平衡增长路径.然而,现有实证研究多聚焦于人力资本对FDI等单一影响的门槛效应的存在性,在经济增长框架下讨论中国人力资本门槛效应的系统研究并不多见.基于此,本文利用中国1996-2012年省级面板数据估计了人力资本门槛值,并考察了人力资本门槛效应对物质资本、FDI以及人力资本等要素回报率的综合影响.研究结果表明人力资本积累的门槛效应确实显著存在,以人均受教育年限计算的门槛值达到9.75年,人力资本水平超过门槛值后物质资本和FDI回报率均大幅提高.     

面板数据的双误差分量模型与Hausman检验

双误差分量模型可以分为四类:固定效应双误差分量模型、随机效应双误差分量模型、个体固定效应-时间随机效应的双误差分量模型、个体随机效应-时间固定效应的双误差分量模型.Hausman检验该如何判断个体或者时间是固定效应还是随机效应呢?文章对此专门进行了研究,给出了一个合理有效的步骤.     

基于中点及半径的区间数组合预测模型性质研究

文章将区间数的中点及半径分别看作实数序列,考虑组合预测区间数的中点及半径序列与实际值中点及半径序列之间的误差平方和,建立多目标最优化模型,通过引入偏好系数将多目标转化为单目标规划问题求解得到组合预测的权重.接着对所建立模型的有效性进行研究,并给出模型的相关性质.通过实例计算验证模型有效性.     

HORVITZ-THOMPSON AND DES RAJ ESTIMATORS REVISITED

In this paper it is shown that the generalized πPS sampling strategy consisting of the design with π_i, the probability of inclusion of the ith unit in the sample, proportional to the modified size together with the corresponding Horvitz-Thompson estimator (Rao, 1971), is superior to the symmetrized Des Raj strategy under a general super-population set-up for all values of the super-population parameter g, when the samples are of size two.     

On the performance of the nearest proportional to size sampling design

Assuming a super-population model the expected variance of the generalized difference estimator (Basu,1971) based on the nearest proportional to size sampling design introduced by Gabler(1987) is shown to be less than that of the same estimator based on an arbitrary sampling design from which the former design is realized. The former strategy is also shown to fare better than an unbiased ratio-cum-generalized difference estimator based on the nearest proportional to size sampling design in the sense of having less expected design variance under the same model.     

Multiple-start balanced modified systematic sampling in the presence of linear trend

ABSTRACT

In this paper, we propose a sampling design termed as multiple-start balanced modified systematic sampling (MBMSS), which involves the supplementation of two or more balanced modified systematic samples, thus permitting us to obtain an unbiased estimate of the associated sampling variance. There are five cases for this design and in the presence of linear trend only one of these cases is optimal. To further improve results for the other cases, we propose an estimator that removes linear trend by applying weights to the first and last sampling units of the selected balanced modified systematic samples and is thus termed as the MBMSS with end corrections (MBMSSEC) estimator. By assuming a linear trend model averaged over a super-population model, we will compare the expected mean square errors (MSEs) of the proposed sample means, to that of simple random sampling (SRS), linear systematic sampling (LSS), stratified random sampling (STR), multiple-start linear systematic sampling (MLSS), and other modified MLSS estimators. As a result, MBMSS is optimal for one of the five possible cases, while the MBMSSEC estimator is preferred for three of the other four cases.     

Some estimation procedures using a linear model in successive sampling

ABSTRACT

In successive sampling some recent works depict the use of super-population models where information on stable auxiliary variable over occasions has been utilized. Stability character of auxiliary variable may not sustain, if the duration between occasions is large. To cope with such situations, the present work is an attempt to develop some estimation procedures by utilizing the information on two independent auxiliary variables through a linear super-population model. Some estimators are proposed to estimate the current population mean in two occasions successive (rotation) sampling. Optimum replacement strategies are formulated and performances of the proposed estimators have been discussed. Results are interpreted through empirical studies.     

An alternative estimator for multiple characteristics in PPS sampling

The problem considered relates to large-scale sample surveys. A new estimator of population total for the characteristics that are poorly correlated with the selection probabilities has been developed for the PPSWR sampling scheme. The relative efficiency of the proposed estimator has been studied under a super-population model. A numerical investigation into the performance of the estimator has also been made.     

An alternative estimator for multiple characteristics in rhg sampling scheme

The problem considered relates to large scale sample surveys, A new estimator of population total for the characteristics that are poorly correlated with. the selection probabilities,has been developed for the RHC sampling scheme, She relative efficiency of the proposed estimator has been studied under a super-population model, A numerical. Investigation into the performance of the estimator, has also been made.     

Estimation under systematic sampling schemes for parabolic populations

In this paper, three sampling-estimating strategies involving linear, balanced and modified systematic sampling are considered for the estimation of a finite population total in the presence of parabolic trend. Using appropriate super-population models, their performances are evaluated. For super-population models with constant variance, Yates corrected estimator under linear systematic sampling is shown to perform well. Choices of variance functions under which modified and balanced systematic sampling perform well are also identified based on extensive numerical studies.     

Circular versions of systematic sampling in the presence of linear trend

In this paper, an alternative approach was employed and derived the explicit expressions of sample mean and its variance for the circular and the diagonal circular systematic schemes since these expressions were unavailable in the existing literature. The study also found the corrected estimator of population mean and average variance under super-population model for these scheme. It further revealed that the circular systematic sampling almost performed better than that of the diagonal circular systematic sampling in both linear and auto-correlated super-population models.     

Synthetic and composite estimation under a superpopulation model

Under a simple superpopulation model for an arbitrary sampling design we derive optimal linear unbiased estimators/predictors of a mean in a domain. They can be viewed as synthetic and composite estimators of small area estimation theory when no auxiliary variable is available. Moreover, we show that the only requirement for optimality of a sampling strategy is to use any sampling plan of fixed sample size together with traditional estimators (as designed for simple random sampling without replacement). Finally, for symmetric sampling plans, simplified formulas (based on the first two moments of sample sizes) for optimal synthetic and composite estimators and their MSE’s are derived. Throughout the paper we consistently use the model-design setup.     

Admissible and optimal sampling strategy for estimating finite population mean in randomized response surveys with multiple responses

ABSTRACT

We consider the problem of estimation of a finite population mean (or proportion) related to a sensitive character under a randomized response model when independent responses are obtained from each sampled individual as many times as he/she is selected in the sample and prove the admissibility of a sampling strategy in a class of comparable linear unbiased strategies. We prove that the admissible strategy is also optimal in this class under a super-population model.