Construction of stochastic volatility model based on realized volatility distribution and its application

• Project/Area Number: 25330047
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: Hiroshima University of Economics
• Principal Investigator: Takaishi Tetsuya 広島経済大学, 経済学部, 教授 (60299279)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Project Status: Completed (Fiscal Year 2015)
• Budget Amount: ¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000); Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2014: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000); Fiscal Year 2013: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
• Keywords: マルコフ連鎖モンテカルロ法 / ハイブリッドモンテカルロ法 / ＳＶモデル / ＧＰＵ計算 / 実現ボラティリティ / OpenACC / SVモデル

Outline of Final Research Achievements

We investigated properties of the realized volatility calculated with various sampling times. The hybrid Monte Carlo algorithm that infers the realized stochastic volatility model was developed. Since the hybrid Monte Carlo algorithm can be parallelized we made a program that executes the GPU calculation. We find that the GPU calculation is faster than the CPU calculation on a personal computer. We also use the OpenACC that enables us to make a GPU program easily, and find that the computational speed of the OpenACC is comparable with that of the CUDA.

Research Products

• [Journal Article] Analysis of Realized Volatility for Nikkei Stock Average on the Tokyo Stock Exchange (2016)
  • Author(s): Tetsuya Takaishi and Toshiaki Watanabe
  • Journal Title: Journal of Physics: Conference Series Volume: 710 Pages: 012010-012010
  • DOI: 10.1088/1742-6596/710/1/012010
  • Peer Reviewed / Open Access / Int'l Joint Research / Acknowledgement Compliant
• [Journal Article] Bayesian Inference of Realized Stochastic Volatility Model and Bias of Realized Volatility (2015)
  • Author(s): Ｔｅｔｓｕｙａ Ｔａｋａｉｓｈｉ
  • Journal Title: International Journal of Engineering Science and Innovative Technology Volume: 4 Pages: 1-7
  • Peer Reviewed / Open Access / Int'l Joint Research / Acknowledgement Compliant
• [Journal Article] GPU Computing in Bayesian Inference of Realized Stochastic Volatility Model (2015)
  • Author(s): Ｔｅｔｓｕｙａ Ｔａｋａｉｓｈｉ
  • Journal Title: Ｊｏｕｒｎａｌ ｏｆ Ｐｈｙｓｉｃｓ： Ｃｏｎｆｅｒｅｎｃｅ Ｓｅｒｉｅｓ Volume: 574 Pages: 012143-012143
  • DOI: 10.1088/1742-6596/574/1/012143
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Journal Article] Bayesian estimation of realized stochastic volatility model by Hybrid Monte Carlo algorithm (2014)
  • Author(s): Tetsuya Takaishi
  • Journal Title: Journal of Physics: Conference Series Volume: 490 Pages: 12092-12092
  • DOI: 10.1088/1742-6596/490/1/012092
  • Peer Reviewed
• [Journal Article] Realized Volatility Analysis in A Spin Model of Financial Markets (2014)
  • Author(s): Tetsuya Takaishi
  • Journal Title: JPS Conference Proceedings Volume: 1 Pages: 19007-19007
  • DOI: 10.7566/jpscp.1.019007
  • Peer Reviewed
• [Presentation] 実現確率的ボラティリティ変動モデルによる株価収益率変動の分析 (2016)
  • Author(s): 高石 哲弥
  • Organizer: 日本金融・証券計量・工学学会
  • Place of Presentation: 慶応義塾大学 三田キャンパス
  • Year and Date: 2016-01-24
• [Presentation] Accelerating the hybrid Monte Carlo algorithm by GPU computing for the realized stochastic volatility model (2015)
  • Author(s): Tetsuya Takaishi and Yubin Liu
  • Organizer: GTC Japan 2015
  • Place of Presentation: 虎ノ門ヒルズフォーラム
  • Year and Date: 2015-09-18
• [Presentation] An application of the hybrid Monte Carlo algorithm for realized stochastic volatility model (2015)
  • Author(s): T.Takaishi, Y.Liu and T.T. Chen
  • Organizer: LATTICE 2015
  • Place of Presentation: Kobe International Conference Center
  • Year and Date: 2015-07-14
  • Int'l Joint Research
• [Presentation] GPGPU Computing in Financial Time Series (2014)
  • Author(s): 高石 哲弥
  • Organizer: GTC Japan 2014
  • Place of Presentation: 東京ミッドタウンホール＆カンファレンス、東京都港区
  • Year and Date: 2014-07-16
• [Presentation] ハイブリッドモンテカルロ法による実現確率的ボラティリティモデルのベイズ推定 (2013)
  • Author(s): 高石 哲弥
  • Organizer: 日本金融・証券計量・工学学会
  • Place of Presentation: 明治大学