Artificial Intelligence, Big Data and Market Efficiency Paradigm : Roadmap and comprehensive guide

Abstract

The way the financial industry works is changing dramatically and quickly thanks to Artificial Intelligence (AI) and Machine Learning (ML). Banks and bankers are turning to these technologies in greater numbers to improve customer service, better manage risk and to streamline many of their internal processes. Therefore, this paper provides an exhaustive review of the extent to which AI and ML have been incorporated into finance to date, their advantages, disadvantages, and the developments in large-scale data processing (Big Data).

The piece focuses on the benefits of adopting these technologies that are technology-driven, ranging from improved efficiency and accuracy of operation, better risk management as well as a superior experience for customers. It also touches on some of the issues and the risks of deploying AI and ML in finance like ethical considerations, data privacy and cybersecurity considerations, and the risk of algorithm bias. Concurrently, the study probes the theoretical and empirical implications of the utilization of AI and Big Data with the EMH. So, this hypothesis which comes right out of neoclassical finance says ‘the price of a financial asset instantly reflects all available information’. By critically examining and summarizing recent journal articles, the study empirically investigates the roles played by machine and deep learning models (i.e., neural networks, SVMs, decision trees and their hybrids) on the predictability of asset prices, trends and volatility. By studying the market from the viewpoint of the classic assumptions of the market efficiency, these advances are reviewed.