Exponential GARCH (EGARCH)
Exponential Generalised Autoregressive Conditional Heteroskedasticity · Also known as: exponential GARCH, Nelson's EGARCH, asymmetric GARCH, EGARCH — Üstel GARCH
EGARCH is an asymmetric GARCH variant, introduced by Nelson in 1991, that models the leverage effect in which bad news raises volatility more than good news of the same size. It captures the negative-shock asymmetry of financial return series by modelling the logarithm of the conditional variance.
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.
+4 more
When to use it
Use EGARCH for financial return series that show volatility clustering and a suspected leverage effect, where downward shocks raise future volatility more than upward shocks. It needs a reasonably long time series (at least about 100 observations) and assumes that ARCH effects have already been detected. It is most appropriate for continuous return data rather than levels, and is preferred over a plain symmetric GARCH when the news impact is asymmetric.
Strengths & limitations
- Captures the asymmetric leverage effect, letting negative shocks raise volatility more than positive shocks of the same size.
- Models the log of the variance, so conditional variance is automatically positive without non-negativity constraints on the parameters.
- Naturally accommodates volatility clustering and persistence common in financial returns.
- Requires a reasonably long financial return series (at least about 100 observations) and prior evidence of ARCH effects.
- Maximum likelihood estimation of the log-variance specification can be numerically sensitive and harder to interpret than a plain GARCH.
- Designed for return series with volatility clustering; it is not a tool for modelling the level of a series or non-financial data without such structure.
Frequently asked
How does EGARCH differ from standard GARCH?
Standard GARCH treats positive and negative shocks of the same size symmetrically and needs non-negativity constraints to keep the variance positive. EGARCH models the log of the variance, which guarantees positivity without constraints, and includes a term that reacts to the sign of the shock, so it captures the asymmetric leverage effect that plain GARCH misses.
What is the leverage effect?
The leverage effect is the empirical tendency for bad news (a negative return) to increase future volatility more than good news (a positive return) of the same magnitude. EGARCH is specifically designed to capture this asymmetry.
How is EGARCH different from GJR-GARCH?
Both capture asymmetric volatility, but EGARCH models the log variance with a sign-dependent term, while GJR-GARCH uses an indicator variable added to a linear variance equation. They are alternative formulations of the same leverage idea.
What data do I need to fit EGARCH?
A continuous financial return series with at least about 100 observations that shows volatility clustering, and ideally prior evidence of an ARCH effect. Heavy tails are common, so a Student-t innovation distribution is frequently used.
Sources
- Nelson, D. B. (1991). Conditional Heteroskedasticity in Asset Returns: A New Approach. Econometrica, 59(2), 347-370. DOI: 10.2307/2938260 ↗
- Engle, R. F. & Ng, V. K. (1993). Measuring and Testing the Impact of News on Volatility. The Journal of Finance, 48(5), 1749-1778. DOI: 10.1111/j.1540-6261.1993.tb05127.x ↗
How to cite this page
ScholarGate. (2026, June 1). Exponential Generalised Autoregressive Conditional Heteroskedasticity. ScholarGate. https://scholargate.app/en/econometrics/egarch
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.
- ARIMAEconometrics↔ compare
- GARCHEconometrics↔ compare
- GJR-GARCHEconometrics↔ compare
- TBATSEconometrics↔ compare