Predicting the Bid-Ask Spread of Equity Options: A Machine Learning Approach Applied to Amazon and AMD

González Pardo, Pablo (2025) Predicting the Bid-Ask Spread of Equity Options: A Machine Learning Approach Applied to Amazon and AMD. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (401kB) | Preview

Abstract

This research investigates the application of machine learning techniques to predict the bid-ask spread of equity options, a key measure of market liquidity and trading costs. While most existing literature focuses on pricing and volatility estimation, this study addresses the topic of direct spread modelling using real world options data from Amazon (AMZN) and AMD, the project develops a full pipeline including feature engineering, exploratory analysis, and the training of both linear and nonlinear models.

Results show that machine learning models significantly outperform traditional linear regressions, especially in capturing nonlinear interactions between variables. Among all models, CatBoost consistently achieves the highest predictive accuracy. Furthermore, explainability tools like SHAP and LIME are used to assess feature importance and enhance model transparency. The findings contribute to both academic understanding and practical applications, providing useful insights for traders, risk analysts, and regulators interested in the behaviour of liquidity in derivatives markets.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Byrne, Brian UNSPECIFIED
Subjects:	H Social Sciences > HG Finance H Social Sciences > HG Finance > Fintech T Technology > T Technology (General) > Information Technology > Fintech Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in FinTech
Depositing User:	Ciara O'Brien
Date Deposited:	24 Jun 2026 10:33
Last Modified:	24 Jun 2026 10:33
URI:	https://norma.ncirl.ie/id/eprint/9392

Actions (login required)

View Item