Asymmetric Power ARCH (APARCH): Flexible Volatility Modelling for Financial Returns
Asymmetric Power ARCH (APARCH) · Also known as: Asymmetric Power ARCH, Power ARCH, APGARCH, Asimetrik Güç ARCH
APARCH, introduced by Ding, Granger, and Engle (1993) while studying long-memory properties of stock market returns, extends the GARCH family by allowing both the power transformation of conditional volatility and an asymmetric response to positive and negative shocks. The model nests at least seven well-known ARCH-type specifications as special cases, making it a unifying framework for volatility modelling in financial econometrics.
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 APARCH when modelling daily or higher-frequency financial return series — equities, foreign exchange, commodities — where volatility clustering, asymmetric shock responses (leverage effect), and uncertainty about the appropriate power transformation are present. Key assumptions are a stationary return process, serial independence in the standardised residuals, and a chosen innovation distribution (normal, Student-t, or skewed-t). APARCH is not appropriate for very short samples (fewer than ~300 observations) or for series without clear ARCH effects. When parsimony is preferred, GJR-GARCH or EGARCH are simpler asymmetric alternatives.
Strengths & limitations
- Nests GARCH, IGARCH, GJR-GARCH, EGARCH, TARCH, and other ARCH variants as special cases, enabling formal likelihood-ratio comparisons
- Simultaneously estimates the optimal power transformation delta and the asymmetry parameter gamma, reducing misspecification risk
- Empirically superior fit in equity and FX markets where the leverage effect is pronounced
- Flexible enough to capture long-memory-like volatility persistence when combined with fractional integration extensions
- Estimation requires numerical optimisation of a nonlinear likelihood, which can suffer from convergence issues or flat regions when the sample is small
- The power parameter delta is estimated with uncertainty and its standard error is often large, complicating inference
- Adding both delta and gamma increases model complexity; the gain over simpler models may be modest in low-volatility series
- Out-of-sample forecasting performance does not always dominate simpler GARCH specifications despite superior in-sample fit
Frequently asked
How does APARCH differ from GJR-GARCH?
GJR-GARCH is a special case of APARCH obtained by fixing the power parameter delta to 2. APARCH additionally estimates delta from the data, allowing the model to determine whether variance, standard deviation, or an intermediate power fits the return series best. This extra flexibility can improve fit but also adds estimation complexity and reduces parsimony.
What software can estimate APARCH models?
APARCH is available in R packages such as rugarch and fGarch, in Stata via the arch command with the power option, in EViews through the GARCH/ARCH dialogue, and in Python via the arch package. Maximum likelihood estimation with robust standard errors is the standard approach; the choice of innovation distribution (normal, t, skewed-t) should be guided by residual diagnostics.
Does a significant leverage parameter always confirm an economic leverage effect?
Not necessarily. A statistically significant gamma indicates that negative shocks increase conditional volatility more than positive shocks of equal size, which is consistent with the financial leverage hypothesis. However, the asymmetry may also reflect other mechanisms such as volatility feedback or investor loss aversion. Residual tests for remaining asymmetry and economic reasoning should complement the statistical finding.
Sources
- Ding, Z., Granger, C. W. J., & Engle, R. F. (1993). A long memory property of stock market returns and a new model. Journal of Empirical Finance, 1(1), 83–106. DOI: 10.1016/0927-5398(93)90006-D ↗
How to cite this page
ScholarGate. (2026, June 2). Asymmetric Power ARCH (APARCH). ScholarGate. https://scholargate.app/en/econometrics/aparch
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 →