限制SUR估计的一个新的Liu型估计

Q3 Mathematics

Journal of Modern Applied Statistical Methods Pub Date : 2020-03-25 DOI:10.22237/JMASM/1556669340

K. Månsson, B. G. Kibria, G. Shukur

{"title":"限制SUR估计的一个新的Liu型估计","authors":"K. Månsson, B. G. Kibria, G. Shukur","doi":"10.22237/JMASM/1556669340","DOIUrl":null,"url":null,"abstract":"A new Liu type of estimator for the seemingly unrelated regression (SUR) models is proposed that may be used when estimating the parameters vector in the presence of multicollinearity if the it is suspected to belong to a linear subspace. The dispersion matrices and the mean squared error (MSE) are derived. The new estimator may have a lower MSE than the traditional estimators. It was shown using simulation techniques the new shrinkage estimator outperforms the commonly used estimators in the presence of multicollinearity.","PeriodicalId":47201,"journal":{"name":"Journal of Modern Applied Statistical Methods","volume":"18 1","pages":"12"},"PeriodicalIF":0.0000,"publicationDate":"2020-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"A New Liu Type of Estimator for the Restricted SUR Estimator\",\"authors\":\"K. Månsson, B. G. Kibria, G. Shukur\",\"doi\":\"10.22237/JMASM/1556669340\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A new Liu type of estimator for the seemingly unrelated regression (SUR) models is proposed that may be used when estimating the parameters vector in the presence of multicollinearity if the it is suspected to belong to a linear subspace. The dispersion matrices and the mean squared error (MSE) are derived. The new estimator may have a lower MSE than the traditional estimators. It was shown using simulation techniques the new shrinkage estimator outperforms the commonly used estimators in the presence of multicollinearity.\",\"PeriodicalId\":47201,\"journal\":{\"name\":\"Journal of Modern Applied Statistical Methods\",\"volume\":\"18 1\",\"pages\":\"12\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-03-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Modern Applied Statistical Methods\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.22237/JMASM/1556669340\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"Mathematics\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Modern Applied Statistical Methods","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.22237/JMASM/1556669340","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Mathematics","Score":null,"Total":0}

引用次数: 1

摘要

提出了一种新的刘氏估计量，用于估计存在多重共线性的参数向量，如果怀疑它属于线性子空间。导出了色散矩阵和均方误差(MSE)。新的估计器可能具有比传统估计器更低的MSE。通过仿真技术表明，在多重共线性情况下，新的收缩估计器优于常用的估计器。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A New Liu Type of Estimator for the Restricted SUR Estimator

A new Liu type of estimator for the seemingly unrelated regression (SUR) models is proposed that may be used when estimating the parameters vector in the presence of multicollinearity if the it is suspected to belong to a linear subspace. The dispersion matrices and the mean squared error (MSE) are derived. The new estimator may have a lower MSE than the traditional estimators. It was shown using simulation techniques the new shrinkage estimator outperforms the commonly used estimators in the presence of multicollinearity.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Modern Applied Statistical Methods STATISTICS & PROBABILITY-

CiteScore

0.50

自引率

0.00%

发文量

期刊介绍： The Journal of Modern Applied Statistical Methods is an independent, peer-reviewed, open access journal designed to provide an outlet for the scholarly works of applied nonparametric or parametric statisticians, data analysts, researchers, classical or modern psychometricians, and quantitative or qualitative methodologists/evaluators.