Bayesian statistical theory based on latent information priors and its applications

2017 Fiscal Year Final Research Report

• Project/Area Number: 26280005
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: The University of Tokyo
• Principal Investigator: Komaki Fumiyasu 東京大学, 大学院情報理工学系研究科, 教授 (70242039)
• Co-Investigator(Kenkyū-buntansha): 諸星 穂積 政策研究大学院大学, 政策研究科, 教授 (10272387) ; 大濱 靖匡 電気通信大学, 大学院情報理工学研究科, 教授 (20243892) ; 村松 正和 電気通信大学, 大学院情報理工学研究科, 教授 (70266071) ; 田中 冬彦 大阪大学, 基礎工学研究科, 准教授 (90456161)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: 予測 / 推定 / ベイズ統計 / 情報量 / 量子統計

Outline of Final Research Achievements

Construction methods for prior distributions in Bayesian analysis have been investigated from theoretical and computational viewpoints. Several theoretical prior construction methods for Bayesian prediction and parameter estimation have been developed. It has been shown that a new metric, which is a generalization of the Fisher metric, plays an essential role. Computational methods for approximating theoretical prior distributions have been proposed. The corresponding results in quantum statistics have also been obtained.

Free Research Field

統計科学