ARCH Model (Autoregressive Conditional Heteroskedasticity)
Autoregressive Conditional Heteroskedasticity Model · Also known as: ARCH, autoregressive conditional heteroskedasticity, Engle ARCH, conditional variance model
The ARCH model, introduced by Robert Engle in 1982, captures time-varying volatility in financial and macroeconomic time series. It models the conditional variance of today's error as a function of past squared errors, explaining why volatile periods cluster together — a phenomenon known as volatility clustering.
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.
+14 more
When to use it
Use ARCH when you have a univariate financial or macro time series — returns, inflation rates, exchange rate changes — and diagnostic tests (ARCH-LM test on OLS residuals) confirm statistically significant heteroskedasticity with volatility clustering. ARCH(q) is most appropriate when the volatility dependence decays quickly over a small number of lags. Do not use ARCH when residuals from the mean equation show no ARCH effects, when the series is not stationary, or when the conditional variance requires many lags — in that case prefer GARCH(1,1), which achieves parsimony.
Strengths & limitations
- Directly models and forecasts time-varying volatility rather than ignoring it.
- Captures volatility clustering, a pervasive feature of financial and macro time series.
- Provides a conditional variance forecast that is indispensable for option pricing, Value-at-Risk, and portfolio optimisation.
- Straightforward maximum likelihood estimation with well-established software support (R, Stata, Python, EViews).
- Earned Robert Engle the 2003 Nobel Prize in Economics, with broad peer-validated application.
- Requires a large number of lags q to capture long-memory volatility, leading to many parameters — GARCH(1,1) is more parsimonious in practice.
- Treats positive and negative shocks symmetrically; asymmetric leverage effects require EGARCH or TGARCH.
- Maximum likelihood estimation can be sensitive to the assumed error distribution; heavy-tailed returns often require a Student-t or GED assumption.
- Does not model multivariate volatility dynamics — use DCC-GARCH for multiple assets.
Frequently asked
How do I decide the ARCH order q?
Run the ARCH-LM (Lagrange Multiplier) test on OLS residuals and inspect the partial autocorrelation function of squared residuals. Choose q where significant autocorrelation disappears. Then compare AIC and BIC across candidate orders and prefer the model with the lowest information criterion. In practice GARCH(1,1) often outperforms ARCH(q) for q > 2.
What is the ARCH-LM test?
The ARCH-LM test regresses squared OLS residuals on their own lags and tests whether the slope coefficients are jointly zero. A significant chi-squared statistic confirms ARCH effects in the residuals, justifying a conditional heteroskedasticity model.
Why prefer GARCH(1,1) over ARCH(q) for large q?
GARCH(1,1) adds one lagged conditional variance term to the ARCH equation. This parsimonious specification mimics an ARCH model with infinitely many lags while estimating only three parameters, and typically fits financial data better than ARCH with many lags.
Can ARCH handle asymmetric volatility responses?
Standard ARCH treats positive and negative shocks of equal size identically because it uses squared errors. Asymmetric leverage effects — where negative shocks raise volatility more than positive ones — require EGARCH (exponential GARCH) or TGARCH (threshold GARCH).
Is Gaussian MLE appropriate for financial returns?
Financial returns often exhibit fat tails, meaning the Gaussian assumption underestimates tail risk. Quasi-maximum likelihood (QML) under Gaussian assumption still gives consistent parameter estimates, but efficiency improves and VaR forecasts become more reliable when a Student-t or GED distribution is assumed.
Sources
- Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica, 50(4), 987–1007. DOI: 10.2307/1912773 ↗
- Engle, R. F. (2001). GARCH 101: The use of ARCH/GARCH models in applied econometrics. Journal of Economic Perspectives, 15(4), 157–168. DOI: 10.1257/jep.15.4.157 ↗
How to cite this page
ScholarGate. (2026, June 3). Autoregressive Conditional Heteroskedasticity Model. ScholarGate. https://scholargate.app/en/econometrics/arch-model
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.
- ARIMA modelEconometrics↔ compare
- DCC-GARCH modelEconometrics↔ compare
- EGARCH modelEconometrics↔ compare
- GARCH ModelEconometrics↔ compare
- TGARCH modelEconometrics↔ compare
- Vector AutoregressionEconometrics↔ compare