Deep Neural Networks for Real-time Market Predictions
Market movement direction predictions should take many financial instruments simultaneously into account due to their correlations. This intrinsict complexity leads to thousands of possible features and thus appropriate for deep neural networks. We apply a deep feed forward network over more than 40 futures based on 5-minute mid-prices. Deep nets substantially outperform traditional models in this setting and based on a reasonable trading strategy backtesting shows increased P&L and Sharpe ratio.
Diego Klabjan is a professor at Northwestern University, Department of Industrial Engineering and Management Sciences. He is also Founding Director, Master of Science in Analytics. His expertise is focused on data science and deep learning with concentration in finance, transportation, and healthcare. Professor Klabjan has led projects with large companies such as The Chicago Mercantile Exchange Group, Intel, General Motors and many other, and he is also assisting numerous start-ups with their analytics needs. He is also a founder of Opex Analytics.