Development of Multivariate Statistical Procedure for High Dimensional and Missing Data and Its Application

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330050
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: Tokyo University of Science
• Principal Investigator: SEO Takashi 東京理科大学, 理学部第一部数理情報科学科, 教授 (00266909)
• Co-Investigator(Renkei-kenkyūsha): NISHIYAMA Takahiro 専修大学, 経営学部, 准教授 (30516472) ; KOIZUMI Kazuyuki 横浜市立大学, 国際総合科学部, 准教授 (70548148) ; SHUTOH Nobumichi 神戸大学, 大学院海事科学研究科, 講師 (00634099) ; HYODO Masashi 大阪府立大学, 工学(系)研究科(研究院), 准教授 (00711764)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 統計科学 / 統計数学 / 統計理論 / 多変量解析法

Outline of Final Research Achievements

On the development of statistical inference theory and test procedure for missing data, in particular, test for mean vectors and simultaneous test for mean vectors and covariance matrices were treated. New modified likelihood ratio test statistics under the general step monotone missing data and multi-sample problem were derived using asymptotic expansion theory. The discriminant analysis under the high dimensional data were considered and obtained some new results. For the multivariate normality test, the results by comparing with previous test procedures under the complete data were given by Monte Carlo simulation.

Free Research Field

数理統計学