A DNN Approach to Improving the Short-Term Investment Criteria for S&P500 Index Stock Market

The investment decision criteria for the S&P500 Index stock market have been a challenge because the data is massive, complex, non-linear and noised. This study proposes the following approach: first, the Index data constitutes the input for training a Deep Learning Neural Network (DNN) for representing it adequately; and second, using the trained model and a sliding window technique, forecast short-term step by step stock values. The process takes into account a heuristic to control the possible extrapolation anomalies of the DNN. Finally, the generated information allows a supported decision. The representation of quantitative correlation values of S&P 500 Index data and their forecasting are promising. Qualitative options of the prediction, constitutes decision making information. The present research permits academics to revise a new experience in data analysis; and, for practitioners, contributes to support investment decisions in the stock market.

[1]  Guohui Zhao,et al.  Deep Learning with Long Short-Term Memory Networks for Air Temperature Predictions , 2019, 2019 International Conference on Artificial Intelligence and Advanced Manufacturing (AIAM).

[2]  Shipra Banik,et al.  Forecasting US NASDAQ stock index values using hybrid forecasting systems , 2015, 2015 18th International Conference on Computer and Information Technology (ICCIT).

[3]  Roded Sharan,et al.  Using deep learning to model the hierarchical structure and function of a cell , 2018, Nature Methods.

[4]  Yan Fang.  Feature Selection, Deep Neural Network and Trend Prediction , 2018 .

[5]  P. J. Haley,et al.  Extrapolation limitations of multilayer feedforward neural networks , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[6]  Gerard de Melo,et al.  Multiple-Weight Recurrent Neural Networks , 2017, IJCAI.

[7]  Shan Gao,et al.  Financial Series Prediction: Comparison Between Precision of Time Series Models and Machine Learning Methods , 2017, ArXiv.

[8]  L.F.A. Wessels,et al.  Extrapolation and interpolation in neural network classifiers , 1992, IEEE Control Systems.

[9]  Morteza Esfandyari,et al.  Stock Market Index Prediction Using Artificial Neural Network , 2016 .

[10]  Zhenyi Yang,et al.  Advantages of Hybrid Deep Learning Frameworks in Applications with Limited Data , 2018, International Journal of Machine Learning and Computing.

[11]  Mahmud Dwi Sulistiyo,et al.  Stock price forecasting using artificial neural network: (Case Study: PT. Telkom Indonesia) , 2017, 2017 5th International Conference on Information and Communication Technology (ICoIC7).

[12]  A. W. Minns,et al.  The extrapolation of artificial neural networks for the modelling of rainfall-runoff relationships , 2005 .

[13]  Ali Osman Pektas,et al.  Investigating the extrapolation performance of neural network models in suspended sediment data , 2017 .

[14]  Magnus Olden.  Predicting Stocks with Machine Learning. Stacked Classifiers and other Learners Applied to the Oslo Stock Exchange , 2016 .

[15]  Aaron Smalter Hall,et al.  Macroeconomic Indicator Forecasting with Deep Neural Networks , 2017 .

[16]  Ritika Singh,et al.  Stock prediction using deep learning , 2016, Multimedia Tools and Applications.

[17]  Samuel Edet,et al.  Recurrent Neural Networks in Forecasting S&P 500 Index , 2017 .

[18]  Ahmad Sharmi Abdullah,et al.  A Stock Market Trading System Using Deep Neural Network , 2017, AsiaSim 2017.

[19]  Tharam S. Dillon,et al.  Computational Intelligence Applications to Option Pricing, Volatility Forecasting and Value at Risk , 2017, Studies in Computational Intelligence.

[20]  Hussein A. Abdou,et al.  Prediction of Financial Strength Ratings Using Machine Learning and Conventional Techniques , 2017 .

[21]  Bing Yang,et al.  Stock market index prediction using deep neural network ensemble , 2017, 2017 36th Chinese Control Conference (CCC).

[22]  Bertrand K. Hassani,et al.  Credit Risk Analysis Using Machine and Deep Learning Models , 2018 .

[23]  Hideki Nakayama,et al.  Deep Learning for Forecasting Stock Returns in the Cross-Section , 2018, PAKDD.