Component GARCH
Component-Based GARCH Model · Also known as: Volatility components model
Component GARCH decomposes conditional variance into transitory (short-term) and permanent (long-term) components with different dynamics, allowing flexibility in capturing volatility behavior at multiple frequencies. Introduced by Engle and Lee (1999), it elegantly models the empirical finding that volatility exhibits both rapid mean-reversion (daily shocks) and slow mean-reversion (level shifts). This framework is crucial for understanding volatility persistence and improving long-horizon volatility forecasting.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use Component GARCH when interested in understanding volatility persistence, forecasting volatility over multiple horizons (daily, weekly, monthly predictions), or explaining why standard GARCH shows high persistence. Valuable for risk management (long-horizon VaR) and macro-finance (permanent vs transitory shocks).
Strengths & limitations
- Captures volatility behavior at multiple time scales within one model
- More parsimonious than allowing full flexibility in shock responses
- Provides economically interpretable permanent and transitory components
- Improves long-horizon volatility forecasts relative to standard GARCH
- Estimation more complex than standard GARCH; convergence can be slow
- Interpretation of permanent versus transitory is not always clear
- Over-parameterization risk if components' dynamics are similar
- Asymptotic inference theory less developed than for single-component GARCH
Frequently asked
What is the difference between permanent and transitory components?
Transitory component exhibits rapid mean-reversion (daily shocks fade in hours/days); permanent component mean-reverts very slowly (if at all) over observed periods. Permanent shocks shift the volatility level for weeks or months.
How many components should I estimate?
Two-component (permanent + transitory) is standard. Three or more is possible but risks over-fitting. Information criteria can guide selection; two components usually suffice.
How do I forecast with Component GARCH?
Forecast each component separately; transitory component reverts quickly, permanent component persists. Combine forecasts at desired horizon. Long-horizon forecasts are primarily permanent-component values.
What if the permanent component appears non-stationary?
If estimated persistence >=1 for permanent component, it is integrated. This is permissible (slow mean-reversion), but verify model stability. If unstable, reconsider specification or include external drivers (macro variables).
Sources
- Engle, R. F., & Lee, G. (1999). A permanent and transitory component model of stock return volatility. Journal of Political Economy, 107(6), 1363-1384. link ↗
- Ling, S., & McAleer, M. (2003). Asymptotic theory and inference for dynamic conditional distribution models. Journal of Econometrics, 106(1), 119-135. link ↗
How to cite this page
ScholarGate. (2026, June 3). Component-Based GARCH Model. ScholarGate. https://scholargate.app/en/econometrics/component-garch
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Causality in Variance TestEconometrics↔ compare
- DCC-MIDASEconometrics↔ compare
- GARCH-MIDASEconometrics↔ compare