ENHANCING TREND FOLLOWING WITH MULTIPLE TECHNICAL SIGNALS by ZHANG JIANWEN Thesis Supervisor, Assistant Professor, SI Yain-Whar (Lawrence) Software Engineering Trend following (TF) is an investment or trading strategy in technical analysis. Investors and traders enter the stock market when they think a trend is established based on their pre-defined rule and follow it, and quit when they believe that a trend is over. Finding the correct trend with proper rules is crucial in TF. One popular way to handle this is by setting a pair of pre-defined thresholds called P&Q, calculated based on the close price of a security. However, TF based on P&Q sometimes not working well for some securities. In this thesis, we proposed an enhancing TF with multiple technical signals to improve the performance of TF based on P&Q. We take advantage of a set of technical indicators such as Simple and Exponential Moving Averages, Moving Average Convergence / Divergence and Relative Strength Index, etc., giving them weights and combining them with TF based on P&Q model to help making decisions on whether it is a good trend or not. A Particle Swarm Optimization algorithm helps us to find the best P&Q thresholds and the weights to each technical signals. Experiments conducted on indices and stocks show promising results.