Financial Time Series Forecasting Using Deep Learning
Abstract: In this work we present a data-driven end-to-end Deep Learning approach for time series prediction, applied to financial time series. A Deep Learning scheme is trained to predict the temporal trends of stocks and ETFs in NYSE or NASDAQ that are part of S&P 500. Our approach is based on a deep neural network (DNN) that is applied to raw financial data inputs, and is trained to predict the temporal trends of stocks and ETFs. In order to handle commission-based trading, we derive an investment strategy that utilizes the probabilistic outputs of the DNN, and optimizes the average return. The proposed scheme is shown to provide statistically significant accurate predictions of financial market trends, and the investment strategy is shown to be highly profitable under this challenging setup. The performance compares favorably with contemporary benchmarks along two-years of back-testing.
* This work was carried out towards the M.Sc. degree in the Faculty of Engineering, Bar-Ilan University, under the supervision of Prof. Yosi Keller.