Developing the methodology of meta-analysis from a unified perspective

2015 Fiscal Year Final Research Report

• Project/Area Number: 24500355
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: The Institute of Statistical Mathematics
• Principal Investigator: Henmi Masayuki 統計数理研究所, データ科学研究系, 准教授 (80465921)
• Co-Investigator(Renkei-kenkyūsha): YOSHIDA Ryo 統計数理研究所, モデリング研究系, 准教授 (70401263) ; YAMASHITA Okito 国際電気通信基礎技術研究所, 脳情報解析研究所, 計算脳イメージング研究室長 (80418516) ; TAKENOUCHI Takashi 公立はこだて未来大学, システム情報科学部, 准教授 (50403340) ; KAWAKITA Masanori 九州大学, 大学院システム情報科学府, 助教 (90435496)
• Research Collaborator: COPAS John ; YONEOKA Daisuke
• Project Period (FY): 2012-04-01 – 2016-03-31
• Keywords: 予測モデル / 回帰モデル / 多変量メタアナリシス

Outline of Final Research Achievements

Meta-analysis is a statistical analysis to obtain stronger evidence by combining statistical results of some studies with the same purpose. In this research, I aimed to develop the methodology of meta-analysis from a unified perspective by considering similar methods to meta-analysis conducted in various research ares as well as medical research, where meta-analysis has been conducted mainly for confirmatory purpose. The main research outcome obtained during the study period includes a method of meta-analysis of regression models for prediction. This method combines regression coefficients by correcting the bias in the case where the sets of covariates (explanatory variables) are different in the studies included in the meta-analysis, and is expected to be applied to many other areas as well as medical research.

Free Research Field

統計科学