Estimating market liquidity from daily data: Marrying microstructure models and machine learning

Yuehao Dai, Chao Shi, Ruixun Zhang,
Estimating market liquidity from daily data: Marrying microstructure models and machine learning,
Journal of Financial Markets,
2025,
101019,
ISSN 1386-4181,
https://doi.org/10.1016/j.finmar.2025.101019.
(https://www.sciencedirect.com/science/article/pii/S138641812500059X)
Abstract: We apply machine learning to estimate daily measures of market liquidity by combining microstructure models with low-frequency daily data only, in stock markets in the United States and China. Boosting trees and neural networks significantly improve the performance across different liquidity measures. Our machine learning models are interpretable and improvements are due to (a) more information from raw data that microstructure models do not capture; and (b) better use of information from learned nonlinear and non-monotonic relationships. We further demonstrate two applications of our trained machine learning models in estimating the illiquidity risk premium and systematic liquidity risk.
Keywords: Liquidity; Bid–ask spread; Microstructure; Machine learning; Interpretability