WEKO3
アイテム
Selection of Variables in Multivariate Regression Models for Large Dimensions
http://hdl.handle.net/2261/33409
http://hdl.handle.net/2261/33409c257a7c9-2349-4483-9ccf-3c99cd71dc53
Item type | テクニカルレポート / Technical Report(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2017-01-17 | |||||
タイトル | ||||||
タイトル | Selection of Variables in Multivariate Regression Models for Large Dimensions | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Akaike information criterion | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Mallows’Cp | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | large dimension | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | multivariate linear regression model | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | selection of variables | |||||
資源タイプ | ||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_18gh | |||||
資源タイプ | technical report | |||||
アクセス権 | ||||||
アクセス権 | metadata only access | |||||
アクセス権URI | http://purl.org/coar/access_right/c_14cb | |||||
著者 |
Srivastava, Muni S.
× Srivastava, Muni S.× Kubokawa, Tatsuya |
|||||
著者所属 | ||||||
値 | Department of Statistics, University of Toronto, CANADA | |||||
著者所属 | ||||||
値 | Faculty of Economics, University of Tokyo, JAPAN | |||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | The Akaike information criterion, AIC, and Mallows'Cp statistic have been proposed for selecting a smaller number of regressor variables in the multivariate regression models with fully unknown covariance matrix. All these criteria are, however, based on the implicit assumption that the sample size is substantially larger than the dimension of the covariance matrix. To obtain a stable estimator of the covariance matrix, it is required that the dimension of the covariance matrix be much smaller than the sample size. When the dimension is close to the sample size, it is necessary to use ridge type of estimators for the covariance matrix. In this paper, we use a ridge type of estimators for the covariance matrix and obtain the modified AIC and modified Cp statistic under the asymptotic theory that both the sample size and the dimension go to infinity. It is numerically shown that these modified procedures perform very well in the sense of selecting the true model in large dimensional cases. | |||||
内容記述 | ||||||
内容記述タイプ | Other | |||||
内容記述 | 本文フィルはリンク先を参照のこと | |||||
書誌情報 |
Discussion paper series. CIRJE-F 巻 CIRJE-F-709, 発行日 2010-01 |
|||||
書誌レコードID | ||||||
収録物識別子タイプ | NCID | |||||
収録物識別子 | AA11450569 | |||||
フォーマット | ||||||
内容記述タイプ | Other | |||||
内容記述 | application/pdf | |||||
日本十進分類法 | ||||||
主題Scheme | NDC | |||||
主題 | 335 | |||||
出版者 | ||||||
出版者 | 日本経済国際共同センター | |||||
出版者別名 | ||||||
値 | Center for International Research on the Japanese Economy | |||||
関係URI | ||||||
識別子タイプ | URI | |||||
関連識別子 | http://www.cirje.e.u-tokyo.ac.jp/research/dp/2010/2010cf709ab.html | |||||
異版あり | ||||||
関連タイプ | hasVersion | |||||
識別子タイプ | URI | |||||
関連識別子 | http://doi.org/10.1080/03610926.2011.624242 |