Market Index and Stock Price Direction Prediction using Machine Learning Techniques: An empirical study on the KOSPI and HSI

The prediction of a stock market direction may serve as an early recommendation system for short-term investors and as an early financial distress warning system for long-term shareholders. In this paper, we propose an empirical study on the Korean and Hong Kong stock market with an integrated machine learning framework that employs Principal Component Analysis (PCA) and Support Vector Machine (SVM). We try to predict the upward or downward direction of stock market index and stock price. In the proposed framework, PCA, as a feature selection method, identifies principal components in the stock market movement and SVM, as a classifier for future stock market movement, processes them along with other economic factors in training and forecasting. We present the results of an extensive empirical study of the proposed method on the Korean composite stock price index (KOSPI) and Hangseng index (HSI), as well as the individual constituents included in the indices. In our experiment, ten years data (from January 1st, 2002 to January 1st, 2012) are collected and schemed by rolling windows to predict one-day-ahead directions. The experimental results show notably high hit ratios in predicting the movements of the individual constituents in the KOSPI and HSI. The results also varify the \textit{co-movement} effect between the Korean (Hong Kong) stock market and the American stock market.