Machine Learning and the Optimal Choice of Asset Pricing Model

Bielinski, Aleksadner; Broby, Daniel

Abstract

This chapter evaluates the traditional methods for price prediction and examines what we believe are the most promising machine learning techniques for that task. Asset price forecasting is one of the fundamental problems in the financial field. Traditional forecasting methods include Capital Asset Pricing Model (CAPM) or Factor Models to estimate stocks’ excess returns. More recently, an increasing number of researchers and financial practitioners began to explore the role of machine learning in asset pricing. We show how these methods have been already applied in practice and discuss their results. We also explore the potential use of neural networks in asset pricing as we believe that their capacity to process large amounts of data together with the ability to accurately capture non-linear relationships among the variables makes them a great tool for price prediction.

Citation

Bielinski, A., & Broby, D. (2023). Machine Learning and the Optimal Choice of Asset Pricing Model. In S. Hasan Jafar, H. K, H. El-Chaarani, S. Moturi, & N. Gupta (Eds.), Artificial Intelligence for Capital Markets (91-127). Taylor & Francis. https://doi.org/10.1201/9781003327745-7

• Acceptance Date: Jun 27, 2022
• Online Publication Date: May 15, 2023
• Publication Date: 2023
• Deposit Date: Jun 30, 2022
• Publicly Available Date: May 16, 2025
• Publisher: Taylor & Francis
• Pages: 91-127
• Edition: 1st
• Book Title: Artificial Intelligence for Capital Markets
• Chapter Number: 7
• ISBN: 9781032353937
• DOI: https://doi.org/10.1201/9781003327745-7
• Public URL: http://researchrepository.napier.ac.uk/Output/2883653