Bootstrapping Time-Varying Uncertainty Intervals for Extreme Daily Return Periods

This study aims to overcome the problem of dimensionality, accurate estimation, and forecasting Value-at-Risk (VaR) Expected Shortfall (ES) uncertainty intervals in high frequency data. A Bayesian bootstrapping backtest density forecasts, which are based on a weighted threshold quantile continuously ranked probability score, developed. Developed backtesting procedures revealed that an estimated Seasonal autoregressive integrated moving average-generalized score-generalized extreme value distribution (SARIMA–GAS–GEVD) with skewed student-t had best prediction performance VaR ES. Extension this non-stationary literature is quite complicated since it requires specifications not only how usual parameters change over time but also those bulk components. implies combination stochastic econometric model theory (EVT) provides robust basis necessary for statistical predictions