Classification-based Financial Markets Prediction using Deep Neural Networks
Deep neural networks (DNNs) are powerful types of artificial neural networks (ANNs) that use several hidden layers. They have recently gained considerable attention in the speech transcription and image recognition community (Krizhevsky et al., 2012) for their superior predictive properties including robustness to overfitting. However their application to algorithmic trading has not been previously researched, partly because of their computational complexity. This talk describes the application of DNNs to predicting financial market movement directions. In particular we describe the configuration and training approach and then demonstrate their application to backtesting a simple trading strategy over 43 different Commodity and FX future mid-prices at 5-minute intervals. All results in this presentation are generated using a C++ implementation on the Intel Xeon Phi co-processor which is 11.4x faster than the serial version and a Python strategy backtesting environment both of which are available as open source code written by the authors.
Matthew Dixon is the CEO and Founder of Quiota LLC- a consulting firm enabling trading firms to adopt machine learning. He is also an Assistant Professor of Finance and Statistics at the Illinois Institute of Technology in Chicago. He began his career as a quantitative developer at Lehman Brothers, and subsequently worked at Barclays Capital and Deutsche Bank. He is a technical advisor and consultant to a number of finance and technology companies in Chicago and is a frequently invited speaker at industry and academic workshops and conferences. Matthew's research has been featured in the Financial Times and has published around twenty peer-reviewed technical articles in the area of applied machine learning and high performance computing. He has held visiting research appointments at Stanford University, UC Davis and collaborates with UC Berkeley and Northwestern University. He chairs the workshop on computational finance at the annual SuperComputing conference and serves on the scientific program committee of HPC'16. Matthew holds a MSc in Parallel and Scientific Computation (with distinction) from the University of Reading (UK), and a PhD in Applied Mathematics from Imperial College London.
|Your Review / Feedback :|
|Your Company :|