S&P 500 Stock Movement Prediction using Deep Learning
Stock market, now 25 Trillion, is one pivotal player in US economy. Predicting stock returns, using either company’s financials or quantitative factors, has been considerd one of the most challenging and rewarding work, due to the noise and volatile features of the time series [1]. In the past decades, both academia and industries have done extensive studies using machine-learning models to predict financial time series, such as support vector machine [2], and neural network [3]. In this project, we would like to attempt to predict stocks returns with cutting-edge technology in deep learning.
This paper is about predicting the movement of stock consist of S&P 500 index. Historically there are many approaches have been tried using various methods to predict the stock movement and being used in the market currently for algorithm trading and alpha generating systems using traditional mathematical approaches [11, 12]. The success of artificial neural network recently created a lot of interest and paved the way to enable prediction using cutting-edge research in the machine learning and deep learning. Some of these papers have done a great job in implementing and explaining benefits of these new technologies. Although most these papers do not go into the complexity of the financial data and mostly utilize single dimension data, still most of these papers were successful in creating the ground for future research in this comparatively new phenomenon.
In this paper, we are trying to use multivariate (not single dimension) raw data instead of engineered matrices data that considers stock split/dividend events (as-is) present in real-world market data. Convolution Neural Network (CNN) so far, the best-known tool for image classification model as of today, is used on the multi-dimensional stock numbers taken from the market mimicking them as vector/matrix of images and the model achieves promising results. The predictions will be done stock by stock, i.e., single stock and not for the portfolio of stocks.