GARCH-MIDAS
GARCH with Mixed Data Sampling · Also known as: Mixed-frequency volatility model
GARCH-MIDAS decomposes volatility into short-term (GARCH) and long-term (MIDAS) components, allowing low-frequency macroeconomic variables to drive medium-term volatility while high-frequency returns govern daily fluctuations. Introduced by Engle and Ghysels (2012), this framework elegantly separates volatility time scales. The approach is powerful for understanding how macro conditions (growth, inflation) drive risk premia and for improved 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 GARCH-MIDAS when studying macro drivers of volatility, forecasting volatility at medium to long horizons, or understanding time-varying risk premia. It is valuable for asset pricing (how macro uncertainty drives expected returns), risk management (macro-conditional VaR), and monetary-policy transmission.
Strengths & limitations
- Naturally separates short- and long-term volatility components
- Incorporates macro information for improved volatility forecasting
- Parsimonious: fewer parameters than full multivariate GARCH
- Economically interpretable decomposition aiding intuition
- Requires specification of macro predictors; misspecification biases results
- Estimation is complex; convergence can be slow
- Assumes linear macro-volatility relationships; nonlinearity requires extensions
- Long-term component often shows slow variation; identification with limited samples can be challenging
Frequently asked
Which macro variables should I include in GARCH-MIDAS?
Theory-guided selection: inflation for equity volatility, credit spreads for credit volatility, VIX expectations for option-implied volatility. Test individual significance; avoid data mining.
How do I decompose volatility into short and long components?
GARCH-MIDAS naturally separates via the model specification: short-term via GARCH dynamics, long-term via MIDAS regression on macro variables. The model outputs both components.
What if macro data is not available at matching frequency?
Use MIDAS polynomial (Almon lag) to handle mixed frequencies directly. For example, monthly macro data predicts daily volatility by weighting months differently in the long-term component.
How do I forecast volatility with GARCH-MIDAS?
Forecast short-term component using standard GARCH recursion; forecast long-term component by projecting macro variables forward. Combine to get total conditional volatility.
Sources
- Engle, R. F., & Ghysels, E. (2012). GARCH for long memory. Journal of Econometrics, 164(2), 385-391. link ↗
- Ghysels, E., Santa-Clara, P., & Valkanov, R. (2005). There is a risk-return trade-off after all. Journal of Financial Economics, 76(3), 674-704. DOI: 10.1016/j.jfineco.2004.03.008 ↗
How to cite this page
ScholarGate. (2026, June 3). GARCH with Mixed Data Sampling. ScholarGate. https://scholargate.app/en/econometrics/garch-midas
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.
Compare side by side →