Machine Learning in Financial Time Series Forecasting: A Systematic Review

Abstract

Time series analysis holds significant theoretical and practical value in the financial field. Due to the complex characteristics of financial time series, such as nonlinearity, dynamics, and chaos, constructing effective prediction models remains a key research direction in both academia and industry. In recent years, with the rapid development of machine learning technology, its application in financial time series prediction achieves remarkable progress. However, most studies remain fragmented and under-reviewed. This paper systematically reviews key research on time series prediction models based on machine learning in the financial field, focusing on analyzing the theoretical modeling and application effects of different models, as well as summarizing the data resources used. It not only compares the performance differences among various models but also discusses the limitations in current prediction modeling processes and proposes possible future improvement directions, aiming to provide references for researchers and practitioners in model selection and optimization. In addition, this paper incorporates the context of computational intelligence and big data, explores the potential value of integrated research approaches, and aims to offer new ideas for advancing the field of financial time series prediction.