Abstract
This study aims to explore the prediction of Taiwan stock price movement and conduct an analysis of its investment performance. Based on Taiwan Stock Market index, the study compares four machine learning models: ANN, SVM, Random Forest and Naïve-Bayes. With a performance evaluation of Taiwan Stock Market index historical data spanning from 2014 to 2018, we find: (1) By overall performance measures, machine learning models outperform benchmark market index. (2) By risk-adjusted measures, the empirical results suggest that ANN generates the best performance, followed by SVM and Random Forest, and Naïve-Bayes coming in last.Keywords: Naive-Bayes, ANN, SVM, Random Forest, Machine Learning, Investment PerformanceJEL Classifications: C11; C53; C63; G11DOI: https://doi.org/10.32479/ijefi.8129
Highlights
In recent years, with the advance of hardware devices, the computing capacity of computers has made a great leap forward, rendering machine learning and artificial intelligence (AI) mainstream
The core objective of the research is to input the results of ten technical analysis indicators into artificial neural networks (ANN), support vector machines (SVM), Random Forest and Naive-Bayes models to predict stock price movement and evaluate investment performance and risk measurement
Among the machine learning models, the results indicate that the order of investment performance excellence can be put down as follows: ANN, SVM, Random Forest and Naïve-Bayes
Summary
With the advance of hardware devices, the computing capacity of computers has made a great leap forward, rendering machine learning and artificial intelligence (AI) mainstream. Stock market price movement prediction has to confront the strongest rejection from the academic paradigm of efficient market hypothesis states that prices of stocks are informationally efficient which means that it is impossible to predict stock prices based on the trading data (Malkiel and Fama, 1970). More recent results show that, if the information obtained from stock prices is pre-processed efficiently and appropriate algorithms are applied trend of stock or stock price index may be predictable (Patel et al, 2015). The new discovery can greatly benefit market practitioners because accurate predictions of movement of stock price indexes are very important for developing effective market trading strategies (Leung et al, 2000)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Economics and Financial Issues
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
