Stock Volatility Forecast Base on Comparative Learning and Autoencoder Framework

Abstract

Volatility is an important indicator of derivatives pricing, financial risk measurement, and market panic sentiment measurement. A reasonable prediction of volatility is of great significance to market participants and regulators. This article proposes a new volatility forecast model. We use comparative learning and autoencoders to improve the accuracy and robustness of the model. Reduce the instability of financial data due to noise. And this article expands traditional machine learning research methods. The traditional model is compared with other deep learning models. Our model has made very competitive progress in accuracy and loss compared to other models.