Specification and estimation of financial processes in the presence of intraday seasonality using noisy high-frequency data

2016 Fiscal Year Final Research Report

• Project/Area Number: 25380266
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Economic statistics
• Research Institution: Konan University
• Principal Investigator: Ishida Isao 甲南大学, 経済学部, 教授 (20361579)
• Project Period (FY): 2013-04-01 – 2017-03-31
• Keywords: ボラティリティ / 日中季節性 / 高頻度データ / 資産価格

Outline of Final Research Achievements

This study empirically demonstrates that the continuous-time GARCH-type stochastic volatility model fits de-seasonalized high-frequency intraday observations of the S&P 500 and Nikkei 225 stock indices better than the Heston-type model. However, specification tests reject both models. In light of these　empirical　results,　this　study　proposes　a functional　HAR　model　for　predicting　future　spot volatility paths, which is a functional version of the HAR model for the realized volatility. This new model treats the dynamic movements of the time-varying seasonal components and the spot volatility in an integrated way, and is found to perform better than the existing approaches in terms of predictive accuracy.

Free Research Field

計量ファイナンス