Investor Confidence and Forecastability of US Stock Market Realized Volatility: Evidence from Machine Learning

Using a machine-learning technique known as random forests, we analyze the role of investor confidence in forecasting monthly aggregate realized stock-market volatility United States (US), over and above wide-array macroeconomic financial variables. We estimate forests on data for period from 2001 to 2020, study horizons up one year by computing forecasts recursive rolling estimation window. find that confidence, especially uncertainty has out-of-sample predictive value overall volatility, well its “good” “bad” variants. Our results have important implications investors policymakers.