The Intelligent project: A machine learning algorithm approach to stock market prediction

• Main Authors: Chua, Ashlee Loraine L., Lam, Jennifer Joy L., Sze, Dana Alyssa L., Tan, Alyanne Haye L.
• Format: text
• Language: English
• Published: Animo Repository 2016
• Subjects: Machine learning; Machine translating; Artificial intelligence; Algorithms; Stock price forecasting
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/8216
• Institution: De La Salle University

In a rapidly innovating environment, the evolution of technology has become fast-paced and more complex. One of these innovations is the artificial intelligence (AI), which enables machines to think and act like humans. The machine learning (ML) algorithms are one group of AI that trains machines to learn certain processes and use this knowledge to deliver output. Because of their innovative state, ML methods are beginning to be used and tested in financial markets. With the growing interest in AI, this research uses ML methods, the Artificial Neural Networks (ANN) and Support Vector Machines (SVM), to predict ASEAN stock indices with with the aid of their correlated indices. These ML approaches are compared with the traditional forecasting methods - ARIMA and linear regression. Following several accuracy tests, results show that generally, ML algorithms generate better forecasts than the traditional methods. Moreover, among the ML methods, the forecasts of the SVM are the most favorable.