Stock Market Prediction Using Elliot Wave Theory and Classification

Abstract

The area of stock market prediction has attracted a great deal of attention during the past decade especially after multiple market crashes. By analyzing market price fluctuations, we can achieve valuable insight regarding future trends. This research proposes a novel method for prediction using pattern analysis and classification. For the first part of the research, a trend analysis algorithm, Elliot wave theory, is used to classify price patterns for DJIA, S&P500, and NASDAQ into three categories: LONG, SHORT, and HOLD. After labeling patterns, classification learning algorithms including decision tree, naïve Bayes, and support vector machine (SVM) are used to learn from the patterns and make a prediction for the future. The algorithm is implemented during the market crashes of May 2010 and August 2015, and the obtained results show that it correctly identifies the market volatility by issuing HOLD and SHORT signals during those crashes.