Abstract
Searching for better and more efficient algorithms to predict stock return and volatility, given the importance of prediction for the stock market return and volatility, is never enough. There are different approaches: econometric, machine learning based and many hybrids between machine learning, neural networks and parametric methods to predict stock market parameters. Almost all of these approaches have demonstrated the superiority of machine learning models in predicting stock volatility. The motivation for this work is the estimation of volatilities on the Warsaw Stock Exchange benchmark price index WIG, which is a relatively less researched stock model, using Support Vector Regression (SVR), Machine Learning (ML) and in particular Deep Learning (DL) approach. The paper also checks these models’ volatility outperformance regarding the traditional and parametric ARCH and GARCH models, using the Root Mean Square Error metric (RMSE). The obtained results are compared to those derived from the New York Stock Exchange Index S&P 500.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have