Stock price forecast with deep learning
This work addresses stock market forecasting for investors and analysts, but it is incremental as it compares existing methods without introducing new ones.
The paper tackled stock price prediction by comparing neural network architectures and optimization techniques, finding that a single-layer recurrent neural network with RMSprop optimizer achieved optimal results with validation and test Mean Absolute Errors of 0.0150 and 0.0148.
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 S&P 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.