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

Statistical inference using auxiliary variables based on optimal transportation

Research Project

Project/Area Number 20K19757
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 60030:Statistical science-related
Research InstitutionHiroshima University

Principal Investigator

Imori Shinpei  広島大学, 先進理工系科学研究科(理), 准教授 (80747345)

Project Period (FY) 2020-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2023: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2022: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2021: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2020: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Keywords補助変数 / 最適輸送理論 / フレシェ距離 / Greedy algorithm / 共変量シフト / ガンマダイバージェンス / Wasserstein距離 / 貪欲法 / 数理統計学 / 変数選択
Outline of Research at the Start

本研究では補助変数を活用した主要変数の解析を考える.すべての補助変数が有用であるとは限らないため,補助変数と主要変数の関連が解析結果にどのように影響するのかを理論的に明らかにすることが肝要である.そこで,変数間の関係性を最適輸送理論に基づき定式化することで,結果に与える影響のメカニズムを解明していく.さらにその結果を用いて,大規模データにも適用可能で理論的に妥当な補助変数の活用手法の構築を目指す.

Outline of Final Research Achievements

Use of auxiliary variables based on the optimal transportation (Wasserstein distance) is studied. An upper evaluation of Wasserstein distance between two mixture distributions in complete data is developed. A property of the Frechet distance when using auxiliary variables and the convergence rate of estimators of the Frechet distance are investigated. Moreover, convergence rates of greedy algorithms under the covariate shift or based on the Gamma divergence are derived.

Academic Significance and Societal Importance of the Research Achievements

有用な補助変数を活用することで主要変数の推定精度の向上が見込まれるため,その理論的な性質の解明は重要である.本研究で導出したWasserstein距離の評価式やフレシェ距離の推定量の収束レートは,これまでの補助変数の活用では考えられていなかったため,今後の補助変数の活用,選択手法の開発の基盤になりうると期待される.また,貪欲法に関する研究内容は高次元データにおける効率的な解析に役立つと考えられる.

Report

(5 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • 2020 Research-status Report
  • Research Products

    (14 results)

All 2024 2023 2022 2021 2020 Other

All Int'l Joint Research (4 results) Journal Article (1 results) (of which Peer Reviewed: 1 results) Presentation (9 results) (of which Int'l Joint Research: 4 results,  Invited: 3 results)

  • [Int'l Joint Research] National Tsing Hua University(台湾)

    • Related Report
      2023 Annual Research Report
  • [Int'l Joint Research] National Tsing Hua University(台湾)

    • Related Report
      2022 Research-status Report
  • [Int'l Joint Research] National Tsing Hua University(台湾)

    • Related Report
      2021 Research-status Report
  • [Int'l Joint Research] National Tsing Hua University(その他の国・地域 台湾)

    • Related Report
      2020 Research-status Report
  • [Journal Article] Asymptotic Optimality of Cp-Type Criteria in High-Dimensional Multivariate Linear Regression Models2023

    • Author(s)
      Imori Shinpei
    • Journal Title

      Statistica Sinica

      Volume: -

    • DOI

      10.5705/ss.202020.0425

    • Related Report
      2021 Research-status Report
    • Peer Reviewed
  • [Presentation] On classification problem based on Frechet distance with auxiliary variables2024

    • Author(s)
      Shinpei Imori
    • Organizer
      The Institute for Mathematical Statistics - Asia-Pacific Rim Meeting (IMS-APRM 2024
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Weighted orthogonal greedy algorithm for prediction under covariate shift2023

    • Author(s)
      Shinpei Imori
    • Organizer
      2023 International Conference for Statistics and Data Science (2023 ICSDS)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Importance weighted orthogonal greedy algorithm with estimated weight function2023

    • Author(s)
      Shinpei Imori
    • Organizer
      The 6th International Conference on Econometrics and Statistics (EcoSta 2023)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Frechet距離に基づく分類と補助変数について2023

    • Author(s)
      伊森晋平
    • Organizer
      研究集会『多変量統計学・統計的モデル選択の新展開』
    • Related Report
      2022 Research-status Report
  • [Presentation] 高次元多変量線形回帰モデルにおける変数選択について2022

    • Author(s)
      伊森晋平
    • Organizer
      日本数学会2022年度秋季総合分科会
    • Related Report
      2022 Research-status Report
    • Invited
  • [Presentation] Frechet距離を用いた分類問題について2022

    • Author(s)
      伊森晋平,若木宏文
    • Organizer
      科研費シンポジウム「多様な分野における統計科学の理論とその応用」
    • Related Report
      2022 Research-status Report
  • [Presentation] 外れ値に対して頑健な貪欲型変数選択手法について2021

    • Author(s)
      伊森 晋平, 橋本 真太郎, Ching-Kang Ing
    • Organizer
      2021 年度統計関連学会連合大会
    • Related Report
      2021 Research-status Report
  • [Presentation] Variable selection in high-dimensional multivariate linear regression models with group structure2021

    • Author(s)
      Shinpei Imori
    • Organizer
      The 4th International Conference on Econometrics and Statistics (EcoSta 2021)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Gamma-divergence に基づく変数選択について2020

    • Author(s)
      伊森 晋平,橋本 真太郎
    • Organizer
      科研費シンポジウム (機械学習・統計学・最適化の数理と AI 技術への展開)
    • Related Report
      2020 Research-status Report

URL: 

Published: 2020-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi