2019/03/11 by Ömer Berat Sezer, Sezer, Omer Berat, Ahmet Murat Özbayoğlu +1
Decision Sciences · Economics, Econometrics and Finance · #FOS: Computer and information sciences #Financial Markets and Investment Strategies #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Stock Market Forecasting Methods
paper · pdf · doi:10.48550/arxiv.1903.04610
openalex publication_date 2019/03/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Even though computational intelligence techniques have been extensively\nutilized in financial trading systems, almost all developed models use the time\nseries data for price prediction or identifying buy-sell points. However, in\nthis study we decided to use 2-D stock bar chart images directly without\nintroducing any additional time series associated with the underlying stock. We\npropose a novel algorithmic trading model CNN-BI (Convolutional Neural Network\nwith Bar Images) using a 2-D Convolutional Neural Network. We generated 2-D\nimages of sliding windows of 30-day bar charts for Dow 30 stocks and trained a\ndeep Convolutional Neural Network (CNN) model for our algorithmic trading\nmodel. We tested our model separately between 2007-2012 and 2012-2017 for\nrepresenting different market conditions. The results indicate that the model\nwas able to outperform Buy and Hold strategy, especially in trendless or bear\nmarkets. Since this is a preliminary study and probably one of the first\nattempts using such an unconventional approach, there is always potential for\nimprovement. Overall, the results are promising and the model might be\nintegrated as part of an ensemble trading model combined with different\nstrategies.\n