• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

High-Dimensional and Large-Sample Asymptotic Theory for the Test Statistics with Monotone Missing Data and Its Application

Research Project

Project/Area Number 16K17642
Research Category

Grant-in-Aid for Young Scientists (B)

Allocation TypeMulti-year Fund
Research Field Foundations of mathematics/Applied mathematics
Research InstitutionKobe University

Principal Investigator

Shutoh Nobumichi  神戸大学, 海事科学研究科, 講師 (00634099)

Project Period (FY) 2016-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2018: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Keywords欠測データ / 仮説検定 / 統計的漸近理論 / 検出力関数 / 平均ベクトル / 非正規母集団 / 漸近検出力 / 欠測値データ / 楕円分布 / 多変量解析 / 高次元データ
Outline of Final Research Achievements

For the statistical inference based on monotone missing data particularly observed by dropout of samples, the test statistics for the missing data mechanism and those for a mean vector were modified via Bartlett-type correction. The asymptotic power functions of the test statistics for a mean vector were also derived. Finally, the simulation results implied that the proposed methods outperformed the existing procedures in aspects of controlling type I error more exactly.

Academic Significance and Societal Importance of the Research Achievements

欠測データ(データセットの一部に欠損があるデータ)はさまざまな研究分野や産業分野においてしばしば観測される.特に,再測定が難しい場面で必要とされる欠測データ解析法において,欠測データメカニズム(標本ベクトルと欠測確率の同時分布の構造)や,母集団分布のパラメータに対する統計的推測は非常に重要である.本研究課題では上記の問題に対する仮説検定を改良し,数値実験により提案手法は標本サイズがそれほど大きくない状況下でも機能する仮説検定であることを確認した.また,仮説検定の性能を表す検出力についても考察を行った.これは応用例が多くみられる欠測データ解析法の改良と性能評価に寄与する結果であると考えられる.

Report

(4 results)
  • 2018 Annual Research Report   Final Research Report ( PDF )
  • 2017 Research-status Report
  • 2016 Research-status Report
  • Research Products

    (10 results)

All 2019 2018 2017 2016 Other

All Journal Article (4 results) (of which Peer Reviewed: 4 results,  Open Access: 1 results,  Acknowledgement Compliant: 1 results) Presentation (5 results) (of which Invited: 1 results) Remarks (1 results)

  • [Journal Article] Asymptotic power comparison of T^2-type test and likelihood ratio test for a mean vector based on two-step monotone missing data2019

    • Author(s)
      Masashi Hyodo, Nobumichi Shutoh
    • Journal Title

      Communications in Statistics - Theory and Methods

      Volume: 印刷中

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] On the asymptotic distribution of T^2-type statistic with two-step monotone missing data2018

    • Author(s)
      Tamae Kawasaki, Nobumichi Shutoh, Takashi Seo
    • Journal Title

      Journal of Statistical Theory and Practice

      Volume: 印刷中 Issue: 3 Pages: 657-668

    • DOI

      10.1080/15598608.2018.1450795

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] Approximation for the percentiles of the sample coefficient of variation2017

    • Author(s)
      Nobumichi Shutoh
    • Journal Title

      Journal of Statistics: Advances in Theory and Applications

      Volume: 18 Issue: 1 Pages: 55-72

    • DOI

      10.18642/jsata_7100121871

    • Related Report
      2017 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Bartlett correction to the likelihood ratio test for MCAR with two-step monotone sample2017

    • Author(s)
      Shutoh, N., Nishiyama, T., Hyodo, M.
    • Journal Title

      Statistica Neerlandica

      Volume: 印刷中

    • Related Report
      2016 Research-status Report
    • Peer Reviewed / Acknowledgement Compliant
  • [Presentation] Effect of nonnormality for the null distribution of T^2-type test statistic based on two-step monotone sample under elliptical distributions2018

    • Author(s)
      首藤 信通
    • Organizer
      2018年度統計関連学会連合大会
    • Related Report
      2018 Annual Research Report
  • [Presentation] 2-step単調欠測データにおける平均ベクトルに関する仮説検定の漸近検出力について2017

    • Author(s)
      小川 颯, 兵頭 昌, 首藤 信通
    • Organizer
      日本計算機統計学会 第31回大会 シンポジウム
    • Related Report
      2017 Research-status Report
  • [Presentation] Bartlett correction to the likelihood ratio test for MCAR with two-step monotone missing data2016

    • Author(s)
      首藤信通,西山貴弘,兵頭昌
    • Organizer
      科研費シンポジウム「統計科学における数理と理論の最先端について」
    • Place of Presentation
      北海道大学大学院経済学研究科(北海道札幌市)
    • Year and Date
      2016-10-29
    • Related Report
      2016 Research-status Report
  • [Presentation] 単調欠測データに基づく線形判別分析と仮説検定法2016

    • Author(s)
      首藤信通
    • Organizer
      2016年日本数学会秋季総合分科会
    • Place of Presentation
      関西大学千里山キャンパス(大阪府吹田市)
    • Year and Date
      2016-09-17
    • Related Report
      2016 Research-status Report
    • Invited
  • [Presentation] Modified likelihood ratio test for MCAR with two-step monotone missing data2016

    • Author(s)
      西山貴弘,首藤信通,兵頭昌
    • Organizer
      2016年度統計関連学会連合大会
    • Place of Presentation
      金沢大学角間キャンパス(石川県金沢市)
    • Year and Date
      2016-09-06
    • Related Report
      2016 Research-status Report
  • [Remarks] 神戸大学研究者紹介システム

    • URL

      http://kuid.ofc.kobe-u.ac.jp/InfoSearch/html/researcher/researcher_wV9RoZ-FRArO-IBQ3mbe3w_ja.html

    • Related Report
      2016 Research-status Report

URL: 

Published: 2016-04-21   Modified: 2020-03-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi