Developing dynamic factor models for economic policy effect and risk assessments

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K03593
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Economic statistics
• Research Institution: Hitotsubashi University
• Principal Investigator: Yamamoto Yohei 一橋大学, 大学院経済学研究科, 教授 (80633916)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: 動学的因子モデル / 構造VAR分析 / 動学的因果効果 / 投機的バブル / ニュース・ショック

Outline of Final Research Achievements

In this research project, I have developed econometric methodologies with a large dimensional panel data set by using dynamic factor models. In particular, the following three topics have been investigated (A) estimating and testing for factor models in the presence of large outliers, (B) incorporating structural changes in dynamic factor models, and (C) constructing valid confidence intervals for the factor process. Besides theoretical developments based on asymptotic theories, several empirical studies have also been implemented. The research outcomes are published as completed working papers and they were presented at several seminars of foreign and domestic universities and international conferences. Multiple papers are accepted by highly ranked academic journals for publication. There are a few technical issues which have not been resolved within the specified period, which will be followed up in my future research.

Free Research Field

計量経済学

Academic Significance and Societal Importance of the Research Achievements

本研究課題で対象とした動学的因子モデルは、昨今のビッグデータ解析の文脈の中で、統計手法の高度化のみならず、経済政策の効果や経済政策上のリスク分析への応用可能性からも学術的および実務的な重要性が高まっている。本課題では、かかる手法を３つの課題に分け、大規模な経済データを用いたマクロ経済政策のリスク分析へ応用するという観点から、先駆的な計量手法を開発した。同時に、それらを用いた実証分析を行った。