Implied volatility directional forecasting: a machine learning approach

This study investigates whether the direction of U.S. implied volatility, VIX index, can be forecast. Multiple forecasts are generated based on standard econometric models, but, more importantly, several machine learning techniques. Their statistical significance is assessed by a plethora performance evaluation measures, while real-time investment strategies devised to appraise implications underlying modeling approaches. The main conclusion analysis that implementation techniques in volatility forecasting effective compared mainstream models and model selection techniques, as they superior both an economic setting.