Can machine learning unlock new insights into high-frequency trading?
By Gbenga Ibikunle, Ben Moews, Khaladdin Rzayev
Rating
1463
Battle Count: 206
Relevance
9/10
Highly relevant for understanding HFT behavior, market microstructure, and the impact of HFT on information acquisition and market quality.
Implementation Complexity
7/10
Requires advanced machine learning techniques and access to large datasets, but uses publicly available data for most of the analysis.
Reproducibility
4/5
The paper provides detailed methodology and uses publicly available data, enhancing reproducibility. However, some proprietary data is used for initial model training.
About this paper
Methodology: Machine Learning for HFT Activity Estimation. Problem types: Regression, Time Series Forecasting.
The interactive Everscope explorer (charts, battles, favorites) loads below.