Liquidity Pricing and Crisis: New Metrics from Traditional and Blockchain-Based Markets

Abstract

This thesis develops novel empirical metrics to advance the understanding of market liquidity, informational efficiency, and financial stability.
Two proposed measures, namely the QV (Quote Volatility) and PM (Price Momentum), are evaluated across multiple asset classes and over diverse temporal regimes, including periods of financial stress and relative calm. These metrics are shown to be informative in contexts such as option illiquidity premia and the early detection and monitoring of financial crises.
Extending the scope to decentralized finance, the thesis introduces the Urgency Score, a wallet-level metric that quantifies the information content embedded in blockchain transaction behavior. This score captures aspects of urgency and informed trading by incorporating features such as transaction timing, gas fees, and counterparty characteristics. To address the inherent nonlinearities in financial markets, machine learning methods are employed throughout the analysis for both prediction and structural inference. These tools enhance the robustness and applicability of the proposed metrics.
The findings offer practical insights for market participants, regulators, and policymakers. The proposed metrics are not only theoretically grounded but also demonstrate empirical relevance, suggesting potential for real-world application in both traditional and blockchain-based financial systems.