Abstract
Bootstrap forecast intervals are developed for volatilities having asymmetric features, which are accounted for by fitting EGARCH models. A Monte-Carlo simulation compares the proposed forecast intervals with those based on GARCH fittings which ignore asymmetry. The comparison reveals substantial advantage of addressing asymmetry through EGARCH fitting over ignoring it as the conventional GARCH forecast. The EGARCH forecast intervals have empirical coverage probabilities closer to the nominal level and/or have shorter average lengths than the GARCH forecast intervals. The finding is also supported by real dataset analysis of Dow–Jones index and financial times stock exchange (FTSE) 100 index.
Original language | English |
---|---|
Pages (from-to) | 1144-1157 |
Number of pages | 14 |
Journal | Communications in Statistics - Theory and Methods |
Volume | 46 |
Issue number | 3 |
DOIs | |
State | Published - 1 Feb 2017 |
Keywords
- Asymmetric volatility
- bootstrapping
- EGARCH model
- volatility forecasting