Bayesian econometric analysis of high-dimensional data

2017 Fiscal Year Final Research Report

• Project/Area Number: 25245035
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Economic statistics
• Research Institution: Kwansei Gakuin University (2015-2017); Kobe University (2013-2014)
• Principal Investigator: KOZUMI Hideo 関西学院大学, 経済学部, 教授 (10261273)
• Co-Investigator(Kenkyū-buntansha): 大森 裕浩 東京大学, 大学院経済学研究科(経済学部), 教授 (60251188) ; 照井 伸彦 東北大学, 経済学研究科, 教授 (50207495) ; 各務 和彦 神戸大学, 経営学研究科, 准教授 (00456005) ; 宮脇 幸治 関西学院大学, 経済学部, 准教授 (40550249) ; 石原 庸博 大阪大学, 金融保険センター, 講師 (60609072)
• Project Period (FY): 2013-10-21 – 2018-03-31
• Keywords: ベイズ統計学 / マルコフ連鎖モンテカルロ法 / 高次元データ

Outline of Final Research Achievements

In this research project, we have developed new Bayesian econometric models for analyzing high-dimensional economic data. Specifically, we have developed Bayesian models based on covariance regression and proposed variable selection methods by applying the stochastic variable selection and shrinkage priors. For the proposed models, we have also developed efficient simulation-based methods. We have investigated the performance of the proposed approaches by conducting simulation studies and real data analysis.

Free Research Field

経済統計学