• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

データ融合による信頼性の高い金融市場モデル構築

Research Project

Project/Area Number 22KJ0544
Project/Area Number (Other) 21J20074 (2021-2022)
Research Category

Grant-in-Aid for JSPS Fellows

Allocation TypeMulti-year Fund (2023)
Single-year Grants (2021-2022)
Section国内
Review Section Basic Section 07060:Money and finance-related
Research InstitutionThe University of Tokyo

Principal Investigator

平野 正徳  東京大学, 大学院工学系研究科, 特別研究員(DC1)

Project Period (FY) 2023-03-08 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥2,500,000 (Direct Cost: ¥2,500,000)
Fiscal Year 2023: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 2022: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 2021: ¥900,000 (Direct Cost: ¥900,000)
Keywordsマルチエージェントシミュレーション / データマイニング / 金融市場 / 人工市場 / 深層学習 / 敵対的生成ネットワーク / 強化学習 / 金融市場シミュレーション / 注文データ
Outline of Research at the Start

金融市場シミュレーションにおいて,シミュレーションの信頼性を向上させるための手法として,データ融合という形で,シミュレーションに実際の市場のデータを導入することを検討,開発する.その結果として,エージェント間での相互関係を取り込むことのできるというメリットのあるマルチエージェントシミュレーションと,実際のデータから特徴を抽出して生かすというデータマイニングの双方のメリットを組み合わせた手法を作成する.

Outline of Annual Research Achievements

近年,金融市場ではデータマイニングとマルチエージェントの活用が進んでいる.しかしながら,その両者の融合が進んでおらず,双方のメリットを生かした技術開発は行われてこなかった.そこで,本研究では,これらの融合技術の開発を行った.まず,金融市場のデータを用いて,PGSGANと呼ばれる,注文生成にサンプリングプロセスを用いた手法を提案し,注文生成において,先行研究より高いパフォーマンスを達成した.そのうえで,その作成したPGSGANを用いて,シミュレーションの定量評価手法を提案し,従来のStylized Factsに基づく,定性的な評価基準と整合性が取れるものであることを示した.さらに,人工市場データマイニングプラットフォームという,人工市場内でデータマイニングの精度評価を行うスキームを提案し,その実例として,金融市場における注文の同時性がデータマイニング手法に与える影響について検証するとともに,このプラットフォームの有効性を示した.また,強化学習手法を活用した,シミュレーションのパラメータチューニング手法について提案し,シミュレーションの定量評価手法と融合させた手法を提案・検証した.この結果,深層学習を用いて,マルチエージェントシミュレーションの妥当性評価と,妥当性のあるパラメータ探索を可能にすることができるということを示すことができた.これらの研究を通じて,シミュレーションとデータマイニングの融合は図ることができた.

Report

(3 results)
  • 2023 Annual Research Report
  • 2022 Annual Research Report
  • 2021 Annual Research Report
  • Research Products

    (19 results)

All 2023 2022 2021

All Journal Article (2 results) (of which Peer Reviewed: 2 results,  Open Access: 1 results) Presentation (16 results) (of which Int'l Joint Research: 10 results) Funded Workshop (1 results)

  • [Journal Article] Neural-network-based parameter tuning for multi-agent simulation using deep reinforcement learning2023

    • Author(s)
      Hirano Masanori、Izumi Kiyoshi
    • Journal Title

      World Wide Web

      Volume: 26 Issue: 5 Pages: 3535-3559

    • DOI

      10.1007/s11280-023-01197-5

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] STBM+: Advanced Stochastic Trading Behavior Model for Financial Markets using Residual Blocks or Transformers2021

    • Author(s)
      Hirano Masanori、Izumi Kiyoshi、Sakaji Hiroki
    • Journal Title

      New Generation Computing

      Volume: 40 Issue: 1 Pages: 7-24

    • DOI

      10.1007/s00354-021-00145-z

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Presentation] Quantitative Evaluation of Multi-agent Simulation using Generative Adversarial Network,2023

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI
    • Organizer
      9th International Conference on Computational Social Science (IC2S2)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Policy Gradient Stock GAN for Realistic Discrete Order Data Generation in Financial Markets,2023

    • Author(s)
      Masanori HIRANO, Hiroki SAKAJI, Kiyoshi IZUMI,
    • Organizer
      13th International Conference on Smart Computing and Artificial Intelligence (SCAI 2023) in 14th IIAI International Congress on Advanced Applied Informatics (IIAI AAI 2023)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] PAMS: Platform for Artificial Market Simulations ~Pythonベースの人工市場シミュレーションプラットフォームと深層学習との融合~2023

    • Author(s)
      平野 正徳, 高田 亮介, 和泉 潔
    • Organizer
      合同エージェントワークショップ&シンポジウム2023 (JAWS2023)
    • Related Report
      2023 Annual Research Report
  • [Presentation] Quantitative Evaluation of Multi-agent Simulation using Generative Adversarial Network -- An Alternative of Qualitative Evaluation for Artificial Market Simulation,2023

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI,
    • Organizer
      The 37th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI2023)
    • Related Report
      2023 Annual Research Report
  • [Presentation] Quantitative Evaluation of Multi-agent Simulation using Generative Adversarial Network -- An Alternative of Qualitative Evaluation for Artificial Market Simulation2023

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI,
    • Organizer
      The 37th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI2023)
    • Related Report
      2022 Annual Research Report
  • [Presentation] Efficient Parameter Tuning for Multi-agent Simulation Using Deep Reinforcement Learning2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI,
    • Organizer
      12th International Conference on Smart Computing and Artificial Intelligence (SCAI 2022-Winter) in 13th IIAI International Congress on Advanced Applied Informatics (IIAI AAI 2022 winter)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Quantitative Tuning of Artificial Market Simulation using Generative Adversarial Network2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI,
    • Organizer
      The 6th IEEE International Conference on Agents (ICA 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Does Order Simultaneity Affect the Data Mining Task in Financial Markets? -- Effect Analysis of Order Simultaneity using Artificial Market2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI,
    • Organizer
      The 24th International Conference on Principles and Practice of Multi-Agent Systems (PRIMA 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Parameter Tuning Method for Multi-agent Simulation using Reinforcement Learning2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI,
    • Organizer
      The 9th International Conference on Behavioral and Social Computing (BESC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Implementation of Actual Data for Artificial Market Simulation2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI, Hiroki SAKAJI,
    • Organizer
      The 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Concept and Practice of Artificial Market Data Mining Platform2022

    • Author(s)
      Masanori HIRANO, Hiroki SAKAJI, Kiyoshi IZUMI,
    • Organizer
      2022 IEEE Symposium on Computational Intelligence for Financial Engineering and Economics (CIFEr)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Data-driven Agent Design for Artificial Market Simulation2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI, Hiroki SAKAJI,
    • Organizer
      The 36th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI2022)
    • Related Report
      2022 Annual Research Report
  • [Presentation] GANを用いた人工市場シミュレーションの定量的評価に向けて2022

    • Author(s)
      平野正徳,和泉潔
    • Organizer
      Symposium on Multi Agent Systems for Harmonization (SMASH) 2022 Summer Symposium
    • Related Report
      2022 Annual Research Report
  • [Presentation] Concept and Practice of Artificial Market Data Mining Platform2022

    • Author(s)
      Masanori HIRANO, Hiroki SAKAJI, and Kiyoshi IZUMI,
    • Organizer
      The 2022 IEEE Computational Intelligence for Financial Engineering and Economics (CIFEr 2022)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Implementation of Actual Data for Artificial Market Simulation2022

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI, and Hiroki SAKAJI,
    • Organizer
      The 21st International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2022)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] STBM+: Advanced Stochastic Trading Behavior Model for Financial Markets based on Residual Blocks or Transformers2021

    • Author(s)
      Masanori HIRANO, Kiyoshi IZUMI, Hiroki SAKAJI
    • Organizer
      The 35th Annual Conference of the Japanese Society for Artificial Intelligence
    • Related Report
      2021 Annual Research Report
  • [Funded Workshop] Special Session on Applied Informatics in Finance and Economics (AIFE) in 14th International Congress on Advanced Applied Informatics2023

    • Related Report
      2022 Annual Research Report

URL: 

Published: 2021-05-27   Modified: 2024-12-25  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi