Gaussian process kernel crossover for automated forex trading system

Due to time varying volatility of forex financial time series price data, the traditional technical analysis such as simple moving average (SMA) and exponential moving average (EMA) cannot capture the dynamic time varying in the market. The combination of Gaussian Process kernel framework and classical trendline crossover strategy is proposed with new objective function for optimization trading performance. The experimental results demonstrate better trading performance for both price prediction and accumulated profit return.