Deployment of a Smart Trading System for Intelligent Stock Trading

In this article we evaluate the deployment of a smart trading system that exploits the features of different technical indicators for intelligent stock trading. Depending on their behaviors, these indicators help in trading under various market conditions. Our smart trading system uses a unified trading strategy that selects five indicators from three well-known categories referred as leading, lagging, and volatility indicators. The trading system looks for common trend signals from at least three indicators within a certain period of time. Collectively generated signals from the technical indicators are used to train a neural network model. The trained neural network model is then used to produce buy and sell signals for trading in stocks. The system is efficient and convenient to use for both individual traders and fund managers. We tested the model on actual data collected from Saudi Stock Exchange and New York Stock Exchange. The performance of the model was checked in terms of percentage returns. The results of the proposed trading model were compared with the benchmark trading strategy. The deployed smart trading system is efficient to produce significant returns over the longer and shorter timeframes.