Big Data Analysis in Finance Management

Abstract

We explore the utilization of machine learning (ML) techniques within the realm of finance research. Initially, we highlight a crucial distinction: supervised and unsupervised learning, the two primary ML categories, tackle distinct challenges compared to the traditional econometric methods. Subsequently, we delve into the contemporary landscape of ML applications in finance, identifying three primary types: (i) the creation of more advanced and innovative metrics, (ii) the minimization of predictive errors, and (iii) the expansion of the conventional econometric toolkit. By organizing these applications into this framework, we provide insight into potential future avenues for both researchers and practitioners. Our findings underscore the numerous advantages of ML methodologies in contrast to traditional approaches, underscoring the immense potential of ML in shaping the future of financial research.