Develop trading strategy combining fundamental analysis, technical analysis and news analysis on SGX stocks under consumer sector

• Main Author: Chan, Vilan
• Other Authors: Wong Jia Yiing, Patricia
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/176854
• Institution: Nanyang Technological University

Over the years, stock price movement prediction is proven to be a very challenging task due to unexpected market behavior and market volatility. In the current work, an effort is made to develop a trading strategy using a combination of Fundamental Analysis, Technical Analysis and News Analysis. Fundamental Analysis aims to sift out fundamentally strong companies that have higher possibility in generating profits for further analysis. Then, the combined results from technical analysis and news analysis will be used as input features to a prediction model. In this project, Logistic Regression and Extreme Gradient Boosting (XGB) classifiers are selected as prediction model to predict daily stock directional movements for stocks in Consumer Sector listed in SGX from January 2018 till December 2022. As trading strategy, the author uses buy/sell prediction signals from the prediction model to simulate a real -life trading environment. The trading strategy is backtested in a self created virtual environment in Python. The result of this study shows that the trading strategy outperformed the passive Buy-Hold strategy in generating profits. 6 out of 8 stocks that are backtested are able to generate better returns than Buy-Hold using this trading strategy and the highest return generated by the trading strategy is 8.36% for Cortina Holdings Limited (C41).