Statistical analysis on high-speed stock price formation with the use of high-frequency limit-order book data

2019 Fiscal Year Final Research Report

• Project/Area Number: 16K03601
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Economic statistics
• Research Institution: Keio University
• Principal Investigator: HAYASHI Takaki 慶應義塾大学, 経営管理研究科(日吉), 教授 (80420826)
• Project Period (FY): 2016-04-01 – 2020-03-31
• Keywords: 高頻度データ / 高頻度トレード (HFT) / マーケット・マイクロストラクチャ / 注文板市場 / 先行遅行分析 / 推薦システム / Kyleモデル / データサイエンス

Outline of Final Research Achievements

The purpose of this study is to obtain empirical knowledge and deepen understanding of stock price formation in the stock markets with increasing speed of execution by statistical analysis of high-frequency limit order book data, and the following results are obtained. (1) An exploratory data analysis of high-frequency lead-lag relationships between stock prices in the domestic stock market is performed and empirical characteristics are demonstrated. (2) Based on wavelet theory, we develop a statistical theory and methodology for estimating the lead-lag times for each of the frequency components of securities price fluctuations. (3) A theoretical model is proposed to explain the trading behaviors of high-speed market participants. (4) A collaborative filtering-based methodology to evaluate the liquidity of individual securities in the high-frequency domain is investigated.

Free Research Field

計量ファイナンス

Academic Significance and Societal Importance of the Research Achievements

(学術) (1)高速での株価の先行遅行関係の実証分析は十分なされておらず, 特に国内は希少であり新規性がある. (2)提案手法は, 複数の先行遅行時間を同時推定できる統計的方法論として新規性・貢献性が特に高い. 理論のベースとなる確率過程モデルは, 不均質市場仮説や, 摩擦のある市場で無裁定性が成立するなどファイナンスの標準理論とも整合的である. (3)高頻度取引業者(HFT)に関する実証的知見と整合的な性質を持つ理論モデルを考案した点で意義がある. (4)推薦モデルの流動性推定問題への適用の点で新規性がある.
(社会) 市場への理解の深化を通じて市場の機能を向上させる可能性がある.