Forecasting Exchange Rates using Time-Varying Econometric Models

2020 Fiscal Year Final Research Report

• Project/Area Number: 17K03809
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Money/ Finance
• Research Institution: Keio University
• Principal Investigator: Ito Mikio 慶應義塾大学, 経済学部(三田), 教授 (70184695)
• Co-Investigator(Kenkyū-buntansha): 野田 顕彦 京都産業大学, 経済学部, 准教授 (80610112) ; 和田 龍磨 慶應義塾大学, 総合政策学部(藤沢), 教授 (20756580)
• Project Period (FY): 2017-04-01 – 2021-03-31
• Keywords: 外国為替市場 / 為替レート予測 / 構造変化推定 / スパース回帰モデル / バンドスペクトラル回帰

Outline of Final Research Achievements

This study examines the possibility of improving foreign exchange rate forecast when one adopts the framework of the time-varying error correction model proposed by Ito et al.(2016) to consider how structural changes affect foreign exchange markets. To attain the goal, we applied the model to three countries' foreign exchange rate data that reflects the comovement of foreign exchange markets. Then, we study how fast the markets return to the long-run state, which is a help of forecast if some deviation occurs. As a result, through the period under study, the speed of the market adjustment was increasing.

Free Research Field

ファイナンス 計量経済学

Academic Significance and Societal Importance of the Research Achievements

外国為替市場は異なる通貨の取引を行う場であり国際的な資金の効率的な配分を本来はもたらす．現実にそうした配分が実現するわけではない．しかし各市場参加者が市場状況に鑑みて自らの利益を高めるような取引を行えば，良く機能する外国為替市場は効率的な資金配分をもたらす．本研究が目指す，外国為替市場の現況の把握，取引者の立場に立つ適切な予測を明らかにすることは国際経済の円滑な運行のために有用であると言える．