Optimal model averaged prediction of counter-factual outcomes for causal inference

Research Project

Project/Area Number	16K17102
Research Category	Grant-in-Aid for Young Scientists (B)
Allocation Type	Multi-year Fund
Research Field	Economic statistics
Research Institution	Kyoto Sangyo University
Principal Investigator	YOSHIMURA Arihiro 京都産業大学, 経済学部, 講師 (40773982)
Project Period (FY)	2016-04-01 – 2018-03-31
Project Status	Completed (Fiscal Year 2017)
Budget Amount *help	¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000) Fiscal Year 2017: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000) Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Keywords	モデル選択 / モデル平均 / 政策評価法 / クロスセクション相関 / モデル選択法 / モデル平均法 / 経済統計学 / 計量経済学
Outline of Final Research Achievements	This study developed an optimal model averaged prediction method of individual counter-factual outcomes for causal inference. The main result of this study is to construct a model averaging method in the presence of cross-sectional dependence in the data and to derive the theoretical properties of the method. Moreover, the finite sample property of the method was investigated by large scale simulation experiments and the better performance of the method relative to the existing methods was found.