Improving Trading Strategies by Optimizing the Weighted Combination of Different Trading Strategies

摘要

This paper proposes a way to increase trading profitability in a market by combining trading strategies through simulated annealing. The profitability of the combination on historical data is used as the objective function for the optimization. We demonstrate our approach by combining five selected trading strategies and testing it on stocks in Stock Exchange of Thailand (SET) although the idea presented here may be used on any set of strategies and any markets. The result shows that the combining strategy performs better than each of the component strategies for both the training data used to optimize and the unseen testing data.