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

A study on asymptotic analysis for robust quasi-posterior distributions

Research Project

Project/Area Number 19K14597
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 12040:Applied mathematics and statistics-related
Research InstitutionTokyo University of Science

Principal Investigator

Nakagawa Tomoyuki  東京理科大学, 理工学部情報科学科, 嘱託特別講師 (70822526)

Project Period (FY) 2019-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2022: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2021: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2020: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2019: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Keywordsベイズ統計 / ロバスト統計 / 漸近理論 / カテゴリカルデータ / 一般化事後分布 / Outlier Rejection / 影響関数 / MCMC / ロバスト推定 / 時系列 / Objective prior / quasi-Bayes / general Bayes / Asymptotic expansion / 擬似ベイズ法 / 漸近論 / 客観ベイズ法
Outline of Research at the Start

近年, 計算機の発達により大規模なデータ多く収集でき, さらに解析が可能になってきた. しかしながら, それらの大量のデータの多くは入力ミスやノイズなどの外れ値というものを多く含むものが多い. 外れ値などが含まれると通常の統計解析の結果は大きく変化してしまい, 妥当性が損なわれる. そのため本研究では外れ値に影響されにくい, 頑健な統計解析手法を開発する必要がある. 本研究では擬似ベイズ法という近年注目を集めている方法に着目し, その手法を用いて頑健な統計解析方法の提案と妥当性を保証する研究を行なっている.

Outline of Final Research Achievements

When dealing with large amounts of data, issues such as outliers and model misspecification can be very serious. Outliers are data points that deviate greatly from the data generating process and can have a significant impact on inference. On the other hand, outliers can also suggest model misspecification, making it important to handle them appropriately. The problem of model misspecification and the presence of outliers has been discussed in Bayesian statistics for a long time. In this study, we developed a Bayesian method that is robust to outliers and investigated its theoretical properties. Specifically, we constructed a robust Bayesian estimator using a pseudo-distance between distributions called divergence and derived its asymptotic properties when performing inference using methods such as MCMC.

Academic Significance and Societal Importance of the Research Achievements

近年は膨大な数のデータが取れるため, その中から外れ値を見つけることは大変困難である. また外れ値に影響されるような手法は, しばしば誤った解析結果を誘導することがある. 一方で外れ値を含む場合は仮定したモデルが誤っている可能性もあるため, 外れ値の扱いは重要である. 特に外れ値に影響を受けにくい解析は, 現在のモデルとデータで説明できる部分の結果を返してくれる. そのため本研究は, データに外れ値が含まれていても影響を受けにくいベイズ法とその理論的性質の導出を行ったことで, 推定だけでなく予測や不確実性の評価も外れ値の影響を受けにくくすることができる.

Report

(5 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • 2019 Research-status Report
  • Research Products

    (22 results)

All 2023 2022 2021 2020 2019 Other

All Int'l Joint Research (1 results) Journal Article (6 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 6 results,  Open Access: 3 results) Presentation (15 results) (of which Int'l Joint Research: 5 results,  Invited: 4 results)

  • [Int'l Joint Research] Institute of Marine Research(ノルウェー)

    • Related Report
      2022 Annual Research Report
  • [Journal Article] An Asymptotic Expansion for the Distribution of Euclidean Distance-Based Discriminant Function in Normal Populations2022

    • Author(s)
      Nakagawa Tomoyuki、Ohtsuka Shuntaro
    • Journal Title

      Journal of Statistical Theory and Practice

      Volume: 16 Issue: 4

    • DOI

      10.1007/s42519-022-00292-6

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Dynamic spatio-temporal zero-inflated Poisson models for predicting capelin distribution in the Barents Sea2022

    • Author(s)
      Sugasawa Shonosuke、Nakagawa Tomoyuki、Solvang Hiroko Kato、Subbey Sam、Alrabeei Salah
    • Journal Title

      Japanese Journal of Statistics and Data Science

      Volume: - Issue: 1 Pages: 1-20

    • DOI

      10.1007/s42081-022-00183-x

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Normalizing transformation of Dempster type statistic in high-dimensional settings2022

    • Author(s)
      Hyodo Masashi、Watanabe Hiroki、Nakagawa Shigekazu、Nakagawa Tomoyuki
    • Journal Title

      Communications in Statistics - Theory and Methods

      Volume: ? Issue: 22 Pages: 1-18

    • DOI

      10.1080/03610926.2022.2056749

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Kick-one-out-based variable selection method for Euclidean distance-based classifier in high-dimensional settings2021

    • Author(s)
      Nakagawa Tomoyuki、Watanabe Hiroki、Hyodo Masashi
    • Journal Title

      Journal of Multivariate Analysis

      Volume: 184 Pages: 104756-104756

    • DOI

      10.1016/j.jmva.2021.104756

    • Related Report
      2021 Research-status Report
    • Peer Reviewed
  • [Journal Article] Two-dimensional index of departure from the symmetry model for square contingency tables with nominal categories2021

    • Author(s)
      T. Momozaki, T. Nakagawa, A. Ishii, Y. Saigusa and S. Tomizawa
    • Journal Title

      Symmetry

      Volume: 13 Issue: 11 Pages: 2031-2031

    • DOI

      10.3390/sym13112031

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] On Default Priors for Robust Bayesian Estimation with Divergences2020

    • Author(s)
      Nakagawa Tomoyuki、Hashimoto Shintaro
    • Journal Title

      Entropy

      Volume: 23 Issue: 1 Pages: 29-29

    • DOI

      10.3390/e23010029

    • Related Report
      2020 Research-status Report
    • Peer Reviewed / Open Access
  • [Presentation] Ordinal response model におけるロバストダイバージェンスを用 いた推定2023

    • Author(s)
      桃﨑智隆, 中川智之
    • Organizer
      RIMS 共同研究『種々の統計的モデルにおける推測方式の有効性』
    • Related Report
      2022 Annual Research Report
    • Invited
  • [Presentation] 外れ値に対する一般化事後分布の頑健性について2022

    • Author(s)
      中川智之, 橋本真太郎, 菅澤翔之助
    • Organizer
      日本計算機統計学会 第 36 回大会
    • Related Report
      2022 Annual Research Report
  • [Presentation] ロバストダイバージェンスを用いたベイズ推論について.2022

    • Author(s)
      中川智之
    • Organizer
      科研費シンポジウム 「大規模複雑データの理論と方法論~新たな発展と関連分野への応用~」
    • Related Report
      2022 Annual Research Report
  • [Presentation] ロバストダイバージェンスを用いたordinal response model に対 するベイズ推定2022

    • Author(s)
      桃﨑智隆, 中川智之
    • Organizer
      2022年度統計関連学会連合大会
    • Related Report
      2022 Annual Research Report
  • [Presentation] ロバストダイバージェンスによる ordinal response model に対す るロバスト推定2022

    • Author(s)
      桃﨑智隆, 中川智之
    • Organizer
      日本計算機統計学会 第36 回大会
    • Related Report
      2022 Annual Research Report
  • [Presentation] ダイバージェンスを用いた ordinal response model におけるロバ ストな推定2022

    • Author(s)
      桃﨑智隆, 中川智之
    • Organizer
      広島大学 統計グループ 金曜セミナー
    • Related Report
      2022 Annual Research Report
    • Invited
  • [Presentation] Comparison of various Robust Bayesian Inference against the outliers.2021

    • Author(s)
      T. Nakagawa
    • Organizer
      Australian and New Zealand Virtual Statistical Conference
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Choice of the Dirichlet parameter for estimation of measures in two-way contingency tables.2021

    • Author(s)
      Tomotaka Momozaki, Koji Cho, T. Nakagawa, Sadao Tomizawa
    • Organizer
      The 11th Conference of the IASC-ARS The Asian Regional Section of the Interna- tional Association for Statistical Computing
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Choice of the Dirichlet parameter to estimate measures for square contingency tables2021

    • Author(s)
      中川智之, 桃﨑智隆, 長光司, 富澤貞男
    • Organizer
      RIMS 共同研究『ベイズ法と統計的推測』
    • Related Report
      2021 Research-status Report
    • Invited
  • [Presentation] ベイズ法を用いた正方分割表における尺度の推定2021

    • Author(s)
      桃﨑智隆, 長光司, 中川智之, 富澤貞男
    • Organizer
      日本計算機統計学会 第 35 回大会
    • Related Report
      2021 Research-status Report
  • [Presentation] Robust Bayesian estimation by using the quasi-posterior with divergence2019

    • Author(s)
      Tomoyuki Nakagawa
    • Organizer
      12th International Conference of the ERCIM WG on Computational and Methodological Statistics
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] Robust Bayesian Inference using γ-divergence2019

    • Author(s)
      Tomoyuki Nakagawa, Shintaro Hashimoto
    • Organizer
      10th International Workshop on Simulation and Statistics
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] ダイバージェンスを用いた外れ値に頑健なベイズ推定2019

    • Author(s)
      中川智之
    • Organizer
      計量経済学ワークショップ
    • Related Report
      2019 Research-status Report
    • Invited
  • [Presentation] Objective Priors in Robust quasi-Bayesian Inference using the Divergences2019

    • Author(s)
      Tomoyuki Nakagawa, Shintaro Hashimoto
    • Organizer
      THE 4TH EASTERN ASIA MEETING ON BAYESIAN STATISTICS, EAC-ISBA 2019
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] γ-divergence を用いたロバストなベイズ推測2019

    • Author(s)
      中川智之, 橋本真太郎
    • Organizer
      応用統計学会2019年年会
    • Related Report
      2019 Research-status Report

URL: 

Published: 2019-04-18   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi