Semiparametric econometric analysis under the existence of endogeneity

2018 Fiscal Year Final Research Report

• Project/Area Number: 15H03335
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Economic statistics
• Research Institution: Kyoto University
• Principal Investigator: Nishiyama Yoshihiko 京都大学, 経済研究所, 教授 (30283378)
• Co-Investigator(Kenkyū-buntansha): 人見 光太郎 京都工芸繊維大学, 基盤科学系, 教授 (00283680) ; 文 世一 京都大学, 経済学研究科, 教授 (40192736) ; 小西 葉子 独立行政法人経済産業研究所, 研究グループ, 上席研究員 (70432060)
• Project Period (FY): 2015-04-01 – 2019-03-31
• Keywords: モデル特定化 / モデル選択 / 内生性 / サービス産業 / ミクロデータ分析

Outline of Final Research Achievements

Endogeneity is an important issue in econometrics using the observed economic data. We obtain the minimax rate of specification test of instrumental variable regression models. We showed that the rate depends on the smoothness of the model. In case of non-smooth model, we showed that the uniform rate is n^{-1/4} and proposed a testing method which attains the rate. As empirical research, we developed an alternative method to measure the values of transport time for freight transportation, which is based on the hedonic model to determine equilibrium freight charge and transport time. We also study other service industries such as retail and wholesale, beauty, and hotel sectors. We implemented empirical analysis by utilizing microdata of government and private sectors to discuss the effects of related policy based on the evidence. We develop unit root tests for AR models under sequential samplings. We derive the joint distribution of the estimated coefficients and the stopping time.

Free Research Field

計量経済学

Academic Significance and Societal Importance of the Research Achievements

計量経済分析において、統計的に扱いの難しい問題に関して、どのように取り扱うのがよいかを調べる研究を行ってきた。特に、構造が大きく変化する可能性がある場合には、どのようなモデルが正しいかを調べるのが難しいことを示した。実証分析においては、トラック輸送、卸売り業、小売業、美容業、宿泊業などの計量分析を行って、それぞれの産業の特色やその産業の従事者の行動に関する経済分析を行った。