Study on estimation and prediction of volatility based on empirical similarity

• Project/Area Number: 18K01554
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 07030:Economic statistics-related
• Research Institution: Kwansei Gakuin University
• Principal Investigator: Morimoto Takayuki 関西学院大学, 理工学部, 教授 (80402543)
• Project Period (FY): 2018-04-01 – 2021-03-31
• Project Status: Completed (Fiscal Year 2020)
• Budget Amount: ¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000); Fiscal Year 2020: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2019: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2018: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
• Keywords: 経験類似度 / 実現測度 / モデル信頼集合 / COVID-19 / 構造変化 / ボラティリティ波及 / 波及ネットワーク / パンデミック / ボラティリティ波及効果

Outline of Final Research Achievements

The achievements of this study mainly consist of the following twofold. (1) Applying the concept of empirical similarity based on case-based decision theory, we quantified the distance between past forecasts and corresponding volatility. Then, we empirically compared the predicted values of volatility obtained from the empirical similarity models with ones from other time-series models. (2) Using industry-specific data in the Japanese financial market before and after the spread of COVID-19 infection, we analyzed structural changes and volatility spillover effects. As a result, we found that structural changes have occurred in many industries in Japan since January 2020.

Academic Significance and Societal Importance of the Research Achievements

(1) 本研究では, 事例ベース意思決定理論に基礎を置いた経験類似度の枠組みに着目し, 経験類似度に基づく時系列モデルを用いボラティリティ予測の実証分析を行なった. ここでの実証研究の中心的貢献は,経験類似度モデルとその他の時系列モデルとの予測力比較にある. 分析結果としては, 外挿予測において経験類似度モデルが最良モデルとして評価された.
(2) COVID‐19 感染拡大前後の日本の金融市場における業種別データを用い, 構造変化とボラティリティ波及効果について分析した. 結果としては, 日本の多くの産業において 2020 年 1 月以降に構造変化が発生していることが示された.

Research Products

• [Journal Article] An Integrated Framework for Visualizing and Forecasting Realized Covariance Matrices (2020)
  • Author(s): H. Shigemoto and T. Morimoto
  • Journal Title: Japanese Journal of Statistics and Data Science Volume: - Issue: 1 Pages: 577-599
  • DOI: 10.1007/s42081-020-00100-0
  • NAID: 210000160487
  • Peer Reviewed / Open Access
• [Journal Article] Analyzing Structural Breaks and Volatility Spillover due to Infectious Disease in Japan: Using Spillover Networks (2020)
  • Author(s): Shigemoto, H. and Morimoto, T.
  • Journal Title: Available at SSRN: https://ssrn.com/abstract=3715379 Volume: -
  • DOI: 10.2139/ssrn.3715379
  • Open Access
• [Journal Article] Dependency Structure Analysis of the Japanese Stock Market Based on Realized Networks (in Japanese: 実現ネットワークによる日本株の依存構造分析) (2020)
  • Author(s): H. Shigemoto and T. Morimoto
  • Journal Title: 日本統計学会誌 Volume: 49 Pages: 241-264
  • Peer Reviewed
• [Journal Article] Incorporating Realized Quarticity into a Realized Stochastic Volatility Model (2019)
  • Author(s): D. B. Nugroho and T. Morimoto
  • Journal Title: Asia-Pacific Financial Markets Volume: 26 Issue: 4 Pages: 495-528
  • DOI: 10.1007/s10690-019-09276-2
  • Peer Reviewed / Int'l Joint Research
• [Presentation] 「A collaborative filtering approach to the evaluation of liquidity in the limit-order book markets」Takaki HAYASHI and Makoto TAKAHASHI へのコメント, (2020)
  • Author(s): T. Morimoto
  • Organizer: 日本ファイナンス学会第28回大会
• [Presentation] Economic Policy Uncertainty and Financial Market Volatility: Evidence from Japan (2019)
  • Author(s): T. Morimoto
  • Organizer: ISI-WSC 2019 (The 62nd International Statistical Institute World Statistics Congress 2019), Kuala Lumpur Convention Centre, Kuala Lumpur, Malaysia
  • Int'l Joint Research
• [Presentation] Multiplicative Error Models with Application to the Japanese Stock Market (2019)
  • Author(s): 森本孝之
  • Organizer: 科学研究費補助金「非ガウス型構造VARモデルの統計理論と応用」計量経済学・計量ファイナンス研究集会