Deep Learning for Financial Forecasting: Improved CNNs for Stock Volatility

Abstract

This study proposes a stock price volatility prediction model based on an improved convolutional neural network (CNN) to improve the accuracy of stock market volatility prediction. By introducing the characteristics of convolutional neural networks, this study can automatically extract multi-level features when processing stock market data and effectively capture the complex patterns in stock price fluctuations. Compared with traditional machine learning models such as support vector machines (SVM) and random forests (RF), and deep learning models such as long short-term memory networks (LSTM), the improved CNN model shows lower mean square error (MSE), mean absolute error (MAE) and root mean square error (RMSE) in the stock price volatility prediction task, showing a more significant advantage. Through experimental verification of European stock market data from 2010 to 2023, the results show that the improved CNN model not only surpasses other comparison models in accuracy but also has strong adaptability and stability. Future research can further explore the combination of other deep learning technologies with CNN to improve the prediction ability of the model while considering the introduction of more external economic factors and multimodal data to provide more accurate decision support for stock market prediction.