Clustering Structure of Microstructure Measures

TL;DR

This paper investigates the clustering structure of market microstructure measures in high-frequency data to identify the most predictive features for stock return prediction, enhancing model accuracy and interpretability.

Contribution

It introduces a clustering model for microstructure measures in 10-second intervals to select the most relevant predictors for stock return forecasting.

Findings

Identifies key microstructure measures for prediction.

Reduces noise by selecting optimal features.

Improves prediction accuracy and interpretability.

Abstract

This paper builds the clustering model of measures of market microstructure features which are popular in predicting stock returns. In a 10-second time-frequency, we study the clustering structure of different measures to find out the best ones for predicting. In this way, we can predict more accurately with a limited number of predictors, which removes the noise and makes the model more interpretable.