How Robust are Limit Order Book Representations under Data Perturbation?
This work addresses data quality challenges for machine learning applications in finance, but it is incremental as it focuses on analysis and guidelines rather than new solutions.
The paper examines the robustness of limit order book representations under data perturbation, identifying issues with existing methods and providing guidelines for future research.
The success of machine learning models in the financial domain is highly reliant on the quality of the data representation. In this paper, we focus on the representation of limit order book data and discuss the opportunities and challenges for learning representations of such data. We also experimentally analyse the issues associated with existing representations and present a guideline for future research in this area.