Parametric estimation for the simple linear regression model under moving extremes ranked set sampling design

Cost effective sampling design is a major concern in some experiments especially when the measurement of the characteristic of interest is costly or painful or time consuming.Ranked set sampling(RSS)was first proposed by McIntyre[1952.A method for unbiased selective sampling,using ranked sets.Australian Journal of Agricultural Research 3,385-390]as an effective way to estimate the pasture mean.In the current paper,a modification of ranked set sampling called moving extremes ranked set sampling(MERSS)is considered for the best linear unbiased estimators(BLUEs)for the simple linear regression model.The BLUEs for this model under MERSS are derived.The BLUEs under MERSS are shown to be markedly more efficient for normal data when compared with the BLUEs under simple random sampling.

A Regression Type Estimator with Two Auxiliary Variables for Two-Phase Sampling

This paper is an extension of Hanif, Hamad and Shahbaz estimator [1] for two-phase sampling. The aim of this paper is to develop a regression type estimator with two auxiliary variables for two-phase sampling when we don’t have any type of information about auxiliary variables at population level. To avoid multi-collinearity, it is assumed that both auxiliary variables have minimum correlation. Mean square error and bias of proposed estimator in two-phase sampling is derived. Mean square error of proposed estimator shows an improvement over other well known estimators under the same case.

Effect of Correlation Level on the Use of Auxiliary Variable in Double Sampling for Regression Estimation

While an auxiliary information in double sampling increases the precision of an estimate and solves the problem of bias caused by non-response in sample survey, the question is that, does the level of correlation between the auxiliary information x and the study variable y ease in the accomplishment of the objectives of using double sampling? In this research, investigation was conducted through empirical study to ascertain the importance of correlation level between the auxiliary variable and the study variable to maximally accomplish the importance of auxiliary variable(s) in double sampling. Based on the Statistics criteria employed, which are minimum variance, coefficient of variation and relative efficiency, it was established that the higher the correlation level between the study and auxiliary variable(s) is, the better the estimator is.

Mixture Regression Estimators Using Multi-Auxiliary Variables and Attributes in Two-Phase Sampling

In this paper, we have developed estimators of finite population mean using Mixture Regression estimators using multi-auxiliary variables and attributes in two-phase sampling and investigated its finite sample properties in full, partial and no information cases. An empirical study using natural data is given to compare the performance of the proposed estimators with the existing estimators that utilizes either auxiliary variables or attributes or both for finite population mean. The Mixture Regression estimators in full information case using multiple auxiliary variables and attributes are more efficient than mean per unit, Regression estimator using one auxiliary variable or attribute, Regression estimator using multiple auxiliary variable or attributes and Mixture Regression estimators in both partial and no information case in two-phase sampling. A Mixture Regression estimator in partial information case is more efficient than Mixture Regression estimators in no information case.

Local Polynomial Regression Estimator of the Finite Population Total under Stratified Random Sampling: A Model-Based Approach

In this paper, auxiliary information is used to determine an estimator of finite population total using nonparametric regression under stratified random sampling. To achieve this, a model-based approach is adopted by making use of the local polynomial regression estimation to predict the nonsampled values of the survey variable y. The performance of the proposed estimator is investigated against some design-based and model-based regression estimators. The simulation experiments show that the resulting estimator exhibits good properties. Generally, good confidence intervals are seen for the nonparametric regression estimators, and use of the proposed estimator leads to relatively smaller values of RE compared to other estimators.

A Modified Regression Estimator for Single Phase Sampling in the Presence of Observational Errors

In this paper, a regression method of estimation has been used to derive the mean estimate of the survey variable using simple random sampling without replacement in the presence of observational errors. Two covariates were used and a case where the observational errors were in both the survey variable and the covariates was considered. The inclusion of observational errors was due to the fact that data collected through surveys are often not free from errors that occur during observation. These errors can occur due to over-reporting, under-reporting, memory failure by the respondents or use of imprecise tools of data collection. The expression of mean squared error (MSE) based on the obtained estimator has been derived to the first degree of approximation. The results of a simulation study show that the derived modified regression mean estimator under observational errors is more efficient than the mean per unit estimator and some other existing estimators. The proposed estimator can therefore be used in estimating a finite population mean, while considering observational errors that may occur during a study.

A network meta-analysis on comparison of invasive and non-invasive sampling methods to characterize intestinal microbiota of birds

Birds maintain complex and intimate associations with a diverse community of microbes in their intestine.Multiple invasive and non-invasive sampling methods are used to characterize these communities to answer a multitude of eco-evolutionary questions related to host-gut microbiome symbioses.However,the comparability of these invasive and non-invasive sampling methods is sparse with contradicting findings.Through performing a network meta-analysis for 13 published bird gut microbiome studies,here we attempt to investigate the comparability of these invasive and non-invasive sampling methods.The two most used non-invasive sampling methods(cloacal swabs and fecal samples)showed significantly different results in alpha diversity and taxonomic relative abundances compared to invasive samples.Overall,non-invasive samples showed decreased alpha diversity compared to intestinal samples,but the alpha diversities of fecal samples were more comparable to the intestinal samples.On the contrary,the cloacal swabs characterized significantly lower alpha diversities than in intestinal samples,but the taxonomic relative abundances acquired from cloacal swabs were similar to the intestinal samples.Phylogenetic status,diet,and domestication degree of host birds also influenced the differences in microbiota characterization between invasive and non-invasive samples.Our results indicate a general pattern in microbiota differences among intestinal mucosal and non-invasive samples across multiple bird taxa,while highlighting the importance of evaluating the appropriateness of the microbiome sampling methods used to answer specific research questions.The overall results also suggest the potential importance of using both fecal and cloacal swab sampling together to properly characterize bird microbiomes.

一个新的CSQR-MIDAS模型及其应用

在经济全球化时代,宏观经济指标受到重大“突发事件”冲击往往会产生异常波动,数据中的异常值会影响模型的预测精度。传统的混频分位数回归QR-MIDAS(Quantile Regression-Mixed Data Sampling)模型的目标函数不可微,且只能给出被解释变量在某个分位数处的预测值。因此,本文将组合预测方法和SQR引入到QR-MIDAS模型中构建了一个新的模型:CSQR-MIDAS模型,通过加权组合不同分位数处预测值能更接近真实值。这将有效解决目标函数不可微的问题,而且在“突发事件”下该模型能更准确地预测宏观经济指标。本文对新冠疫情背景下的美国工业生产指数增长率进行实证分析,结果显示CSQR-MIDAS模型比QR-MIDAS模型的预测误差更低,前者预测误差低于80%分位数下的SQR-MIDAS模型,验证了CSQR-MIDAS模型在“突发事件”下具有稳健性和更优的预测能力。

Mixture Regression-Cum-Ratio Estimator Using Multi-Auxiliary Variables and Attributes in Single-Phase Sampling

In this paper, we have proposed a class of mixture regression-cum-ratio estimator for estimating population mean by using information on multiple auxiliary variables and attributes simultaneously in single-phase sampling and analyzed the properties of the estimator. An empirical was carried out to compare the performance of the proposed estimator with the existing estimators of finite population mean using simulated population. It was found that the mixture regression-cum-ratio estimator was more efficient than ratio and regression estimators using one auxiliary variable and attribute, ratio and regression estimators using multiple auxiliary variables and attributes and regression-cum-ratio estimators using multiple auxiliary variables and attributes in single-phase sampling for finite population.

基于MIDAS模型的中国股市对居民消费的影响效应

根据混频数据抽样模型,实证研究了中国股市对居民消费的影响效应,深入讨论了牛市和熊市阶段股市收益和波动对居民消费的影响特征。已有研究都是使用同频数据,多数研究认为股市对居民消费的影响不显著或不稳定。而本文基于混频数据的分析却得出不同的结论:不论是股市收益还是股市波动均对居民消费有着显著的影响效应。通常股市收益对居民消费有正的影响效应且影响持续时间长,而股市波动对居民消费有负的影响效应且持续性很短。股市收益在牛市阶段具有较大的影响;相反,股市波动在熊市阶段具有较大的影响。

基于最小二乘回归的定量指标一致性评价的样本量公式研制

目的构建并验证基于最小二乘回归法的定量资料一致性评价的样本量计算公式。方法根据最小二乘回归法定量一致性评价的原理,结合统计推断和公式推导,获得基于回归常数和回归系数的样本量计算公式。同时,使用3个实例数据验证公式的准确性,并与Bland-Altman(BA)法样本量计算公式结果进行对比。结果通过公式推导得到回归法定量一致性评价的样本量计算公式,并用实例验证了本公式的准确性。此外,本公式与BA法样本量公式所得结果具有一定差异,在回归法样本量公式计算结果下,回归分析与BA法分析均能得到一致性结论;而在BA法公式结果下,回归分析与BA法分析一致性结论有时不成立。结论本研究首次提出用于回归法定量一致性评价的样本量公式,为该领域研究提供了方法学支持。

山东烟田烟蚜种群动态与空间分布

烟蚜在山东烟田发生普遍,危害严重。研究山东烟田烟蚜种群动态与空间分布对于提高山东烟田烟蚜预测测报准确率以及有效防控该害虫具有重要意义。试验开展了烟蚜的田间种群消长动态调查,并应用5个聚集度指标(丛生指数I,聚块指标m^(*)/m,久野指标Ca,扩散系数C,负二项分布K值)和2种回归方法(Iwao回归分析法与Toylar幂法则)研究其空间分布特征。结果表明,2020—2021年烟田烟蚜的种群动态均为双峰型曲线:第1个高峰出现在5月下旬至6月上旬,第2个高峰出现在7月上中旬。在垂直分布上,烟草植株上部叶片的烟蚜数量占比(49.07%~70.29%)显著高于中部叶片(27.64%~33.71%)和下部叶片(1.64%~19.85%)。数据分析结果显示,I(1.733 2~42.703 0)>0,m^(*)/m(2.368 3~10.414 2)>1,Ca(1.368 3~9.414 2)>0,C(2.733 2~43.703 0)>1,0<K(0.106 2~0.730 8)<8;在Iwao回归方程中,β(4.578 65)>1,α(0.190 57)>0,在Toylar幂法则回归方程中,lga(0.691 65)>0,b(1.817 05)>1,说明调查期间烟蚜种群呈聚集分布,分布的基本成分是个体群。利用空间分布参数确定了烟蚜的理论抽样数,并提出合理的田间抽样技术,即在烟田内采用“Z”字形取样法,每点选取烟草5~10株、每株烟草调查上中部叶片2~4片。

基于混频数据抽样的已实现EGARCH模型的波动率预测

该文以沪深300股指期货高频数据为样本,在Realized EGARCH模型的基础上引入了混频数据抽样(MIDAS)结构与时变波动,构建了基于偏t分布的SMA-Realized EGARCH MIDAS模型,该模型提高了模型捕捉长记忆性的能力,更好地刻画了模型的时变波动性.通过滚动时间窗的方法对模型进行VaR预测与后验测试,采用MCS检验评估各模型在不同测度下的波动率预测能力.研究结果显示:相比于传统的Realized GARCH模型、Realized EGARCH模型和Realized EGARCH MIDAS模型,本文提出的SMA-Realized EGARCH MIDAS模型具有更好的样本拟合效果与样本外波动率预测精度.

美国不确定性冲击对全球股市波动的影响研究

选取2000—2021年美国经济、金融、经济政策和地缘风险四类不确定性指数以及全球GDP排行前15国家股市收益率数据,基于多维不确定性冲击框架,运用单因子、双因子和多因子混频波动率GARCH-MIDAS模型,从样本内拟合与样本外预测两个方面实证考察美国不确定性冲击对全球主要国家股市波动的差异化影响。研究表明:美国经济不确定性和金融不确定性对多数国家股市长期波动均有正向推动作用,其中,美国金融不确定性的影响最为广泛;美国金融不确定性是影响中国股市长期波动的主要因素,且中美贸易摩擦主要通过美国金融不确定性传导;美国经济政策不确定性上升会增加俄罗斯和墨西哥股市长期波动,美国地缘风险对意大利股市长期波动存在显著正向影响。

基于支持向量回归和分位数的雷达K分布海杂波形状参数估计方法

针对传统的雷达K分布海杂波形状参数估计方法在异常样本存在情况下估计精度严重下降的问题,该文提出一种基于支持向量回归(SVR)和样本分位数比值的K分布海杂波形状参数估计方法。首先给定K分布杂波参数和分位数位置的值,根据K分布的累积分布函数计算样本分位数比值及其对数,然后建立以样本分位数比值对数为输入、待估计形状参数为输出的SVR模型,通过交叉验证确定SVR模型的超参数,最后训练SVR模型实现对K分布海杂波形状参数的稳健精确估计。仿真和实测雷达数据表明,所提方法的估计误差低于基于矩的估计方法的估计误差,并且与基于分位数的估计方法具有相近估计性能。此外,相比已有基于分位数的方法,所提方法的超参数容易确定,并且不依赖于查表。

基于强度折减采样与高斯过程回归的空间变异土坡自适应可靠度分析

土体参数空间变异性对土坡稳定性的影响已受到岩土工程界的广泛关注,繁冗的计算量是空间变异土坡可靠度分析面临的瓶颈问题。基于强度折减采样(SRS)和高斯过程回归(GPR)模型,提出一种适用于高维空间的空间变异土坡自适应可靠度分析方法(SRS-GPR)。首先采用Karhunen-Loève展开法将空间变异土体强度参数离散为高维随机变量,随后根据SRS生成土坡临界样本点,接着通过GPR构建土体参数随机场与边坡安全系数之间的高维非线性函数关系,并基于主动学习策略自动搜寻最优训练样本点,迭代更新GPR模型。在此基础上,结合GPR模型和蒙特卡洛模拟进行边坡可靠度分析。最后,通过两个典型算例验证所提方法的准确性、高效性、鲁棒性和适用性。结果表明:所提方法可有效识别靠近极限状态面附近的最优样本点,使得迭代更新的GPR模型在该区域的预测精度逐渐提高。此外,所提方法不受随机变量维度的影响,可直接在高维参数空间应用,且对边坡稳定性模型的调用次数较少,在计算效率方面具有显著优势。

基于特征间关系合成少数类样本的过采样算法

数据不平衡的现象在现实生活中非常普遍。为了提高整体分类精度,分类器有时会以错分少数类为代价。但在现实生活中,对少数类进行错误分类的后果非常严重。考虑到传统重采样算法容易忽略数据的空间分布和少数类样本特征之间的关系,提出一种基于特征关系的采样算法(SABRF)生成新的样本集。SABRF通过帕累托多目标特征选择保留不平衡数据集的关键区分特征,同时通过极端梯度提升(XGBoost)回归模型捕获少数类样本关键特征之间的关系。此外,还提出一个新的样本选择策略衡量新生成样本的质量。使用6个公开的UCI数据集和1个真实的骨科术后血栓数据集进行实验,结果表明,SABRF在受试者工作特征曲线下面积(AUC)、F1分数(F1_score)和几何平均值(G_mean)上均有较好的表现;此外,对使用基于多指标评价的样本选择策略挑选出的新样本进行分类,不平衡数据的分类结果也最好,验证了样本选择策略的有效性。

部分线性变系数模型的贝叶斯分位数回归

针对部分线性变系数模型的参数估计问题,采用B样条方法逼近非参数部分的未知光滑函数,进而利用非对称拉普拉斯分布实现贝叶斯分位数回归,并基于Gibbs抽样推导出所有未知参数的后验分布,通过数值模拟比较分析了贝叶斯分位数回归与分位数回归参数估计的优劣,结果表明,在均方误差准则下,贝叶斯分位数回归的估计效果更优。最后,通过实例分析说明了所提方法的有效性。

粒子群算法优化的广义回归神经网络求解流形学习样本外点问题

目前流形学习已成功应用于降维和数据可视化领域,但在监督分类中的应用效果并不理想,解决好样本外点问题对其应用效果至关重要。基于此,采用粒子群算法优化广义回归神经网络计算测试样本的低维嵌入,获得的结果可直接用于分类。借助粒子群算法的全局搜索能力对处理样本外点问题具有较好的预测性能;在使用糖尿病、虹膜和声呐三个公开数据集的实验中,粒子群算法优化广义回归神经网络的分类总体精度分别为77.63%、100%和88.89%,优于其他8种分类方法,表明该算法可行、有效;同时,该算法能显著降低数据复杂度,提高了预测、模式分类和机器学习的准确性。

基于极端冲击的GARCH-MIDAS模型对股市波动率预测研究

金融危机和政治事件等重大事件会对股市产生极端冲击,为探究极端冲击和非对称效应对股市波动率预测的影响,对GARCH-MIDAS模型进行改进并探究改进后的模型能否提高股票波动率的预测性能。以上证综指数据样本为例,运用滚动时间窗的样本外预测方法和MCS检验,探究同时包含极端冲击和非对称效应的GARCH-MIDAS模型的预测性能。样本内结果表明,我国股市存在明显的杠杆效应和极端冲击且负极端冲击的影响大于正极端冲击;样本外结果表明,在不同的损失函数条件下,MCS检验的结果显示在长期项和短期项中均考虑极端冲击和非对称效应的GARCH-MIDAS模型的预测性能更好。这说明在股票波动率预测模型中考虑极端冲击和非对称效应能够提高模型的预测精度。