Dynamical properties of volume at the spread in the Bitcoin/USD market

Paper: arXiv 2304.01907

Authors: Roberto Mota Navarro, Francois Leyvraz, Hernán Larralde

Abstract

The study of order volumes in financial markets has shown that these display several non-trivial statistical properties. Most studies have been focused on the bulk properties of volume of incoming orders or of realized transactions rather than the dynamical aspects. The present work is a study of the dynamical properties of volume. Unlike previous works, we studied the volume available at the spread rather than the volume of incoming orders or of realized transactions. We found evidence that suggests mean reverting volume changes and strong asymmetries in the equilibrium of sell and buy orders as well as the presence of clustering.

Complexity vs Empirical Score

• Math Complexity: 5.5/10
• Empirical Rigor: 6.5/10
• Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper investigates the dynamic properties of order book volume specifically at the bid-ask spread in the BTC/USD market, offering novel insights into market microstructure. It combines solid statistical analysis of high-frequency data with moderate mathematical modeling of mean reversion and order clustering.

Research Flowchart

  flowchart TD
    A[Research Goal: Analyze dynamical properties of volume at the spread in BTC/USD] --> B[Data Input: Bitcoin/USD Limit Order Book Market Data]
    B --> C[Methodology: Track Volume Changes & Imbalances at Bid-Ask Spread]
    C --> D[Computational Analysis: Measure Reversions, Asymmetries & Clustering]
    D --> E[Key Finding: Mean Reverting Volume Changes]
    D --> F[Key Finding: Buy/Sell Equilibrium Asymmetries]
    D --> G[Key Finding: Order Imbalance Clustering]