KEYWORDS: Data modeling, Visualization, Convolution, Neural networks, Time series analysis, Performance modeling, Machine learning, Image classification, Solids, Social sciences
Stock price prediction has always been an issue full of challenges due to its complex volatile nature. Inspired by the observation that the traders prefer judging the stock trends from the charts, and have concluded useful techniques for predicting stock trends. In this paper, we try to predict stock price trends based on the stock images by using Deep Convolutional Neural Networks (DCNN), and mainly focus on how to generate the stock images and label the image to train the DCNN model. Specifically, on one hand, we transfer the stock time series data to 5 types of stock images, trying to explore effective visual representation for stock data. On the other hand, we label the images in a more meaningful way to train the attention-based DCNN model to build a practical model. Experimental results on the S&P 500 stocks show that the MACD image performs better than other types, and it is hard to predict the stocks’ next day trend. However, the convolutional block attention module (CBAM) can improve the performance of DCNN.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.