Comparison of the Performance of Different Machine Learning Methods in Predicting VIX Volatility

Abstract

As a matter of fact, index volatility has always been one of the key indicators of the state of an index and a reflection of investor confidence and expectations in the market. Among various indicators, the VIX, which is also known as the "Panic Index", has always been viewed by the market as a barometer of the state of the economy. With this in mind, the purpose of this study is to investigate the process of Random Forest, Support Vector Regression as well as XGBoost in predicting VIX volatility and to evaluate their performance. Based on the evaluations, experiments in this study show that XGBoost performs optimally for smaller, low-dimensional time series data. According to the analysis, this study strongly demonstrates the excellent performance as well as great potential of machine learning algorithms in volatility prediction. Overall, these results will provide a feasible solution for volatility prediction with machine learning algorithms.