Estimating Market Liquidity from Daily Data: Marrying Microstructure Models and Machine Learning

51 Pages Posted: 3 Mar 2023 Last revised: 19 Dec 2023

See all articles by Yuehao Dai

Yuehao Dai

Peking University

Ruixun Zhang

Peking University; MIT Laboratory for Financial Engineering

Date Written: December 9, 2022

Abstract

We apply machine learning to estimate the average daily bid-ask spread by combining classical microstructure models with widely available low-frequency (daily) data, in the US and Chinese stock markets. Boosting trees and neural networks significantly improve performance, particularly in terms of cross-sectional correlations and in the Chinese market, both of which address major challenges of microstructure models. 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 utilization of information from learned nonlinear and non-monotone relationships, allowing microstructure models to contribute only when relevant.

Keywords: Liquidity; Bid-ask spread; Microstructure; Machine learning; Interpretability

JEL Classification: C45, G12, G14, G15

Suggested Citation

Dai, Yuehao and Zhang, Ruixun, Estimating Market Liquidity from Daily Data: Marrying Microstructure Models and Machine Learning (December 9, 2022). Available at SSRN: https://ssrn.com/abstract=4371650 or http://dx.doi.org/10.2139/ssrn.4371650

Yuehao Dai

Peking University ( email )

No. 38 Xueyuan Road
Haidian District
Beijing, Beijing 100871
China

Ruixun Zhang (Contact Author)

Peking University ( email )

5 Yiheyuan Road
Haidian District
Beijing, Beijing 100871
China

MIT Laboratory for Financial Engineering ( email )

100 Main Street
E62-611
Cambridge, MA 02142

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
238
Abstract Views
1,001
Rank
240,883
PlumX Metrics