Stock price forecast with deep learning
Institute of Electrical and Electronics Engineers Inc.
In this paper, we compare various approaches to stock price prediction using neural networks. We analyze the performance fully connected, convolutional, and recurrent architectures in predicting the next day value of SP 500 index based on its previous values. We further expand our analysis by including three different optimization techniques: Stochastic Gradient Descent, Root Mean Square Propagation, and Adaptive Moment Estimation. The numerical experiments reveal that a single layer recurrent neural network with RMSprop optimizer produces optimal results with validation and test Mean Absolute Error of 0.0150 and 0.0148 respectively. © 2020 IEEE.
This conference paper is not available at CUD collection. The version of scholarly record of this conference paper is published in 2020 International Conference on Decision Aid Sciences and Application (DASA) (2020), available online at: https://doi.org/10.1109/DASA51403.2020.9317260
convolutional neurons, deep learning, recurrent neurons, SP 500 prediction, time-series forecasting, Backpropagation, Decision support systems, Electronic trading, Financial markets, Forecasting, Gradient methods, Multilayer neural networks, Network layers, Optimization, Stochastic systems, Mean absolute error, Moment estimation, Numerical experiments, Optimization techniques, Root Mean Square, Stochastic gradient descent, Stock price forecasts, Stock price prediction, Recurrent neural networks
Kamalov, F., Smail, L., & Gurrib, I. (2020, November). Stock price forecast with deep learning. In 2020 International Conference on Decision Aid Sciences and Application (DASA) (pp. 1098-1102). IEEE. https://doi.org/10.1109/DASA51403.2020.9317260