Time series analysis and forecasting in finance: A data mining approach
Afiz Adewale Lawal, Omogbolahan Alli, Aishat Oluwatoyin Olatunji, Enuma Ezeife, Ehizele Dean Okoduwa · Open Access Research Journal of Science and Technology · 2023
Time series analysis and forecasting are essential methodologies in finance, playing a pivotal role in predicting market trends, evaluating economic conditions, and supporting decision-making. These methods rely on analyzing sequential data to uncover patterns, trends, and seasonal variations that drive financial phenomena. Traditional statistical models, such as ARIMA and GARCH, have long been utilized; however, their effectiveness is often constrained by assumptions like linearity and stationarity. Recent advancements in data mining techniques, including machine learning and artificial intelligence, have transformed the landscape of time series forecasting. These innovative approaches excel at handling non-linear relationships, high-dimensional data, and noise inherent in financial markets, making them indispensable for modern financial analytics. This paper scours the concept of time series analysis and data mining, examining their integration to improve forecasting accuracy. Additionally, it evaluates challenges such as data quality and computational requirements, while highlighting emerging opportunities, such as real-time forecasting and big data applications.