In econometrics, the autoregressive conditional heteroskedasticity (ARCH) model is a statistical model for time series data that describes the variance of the current error term or innovation as a function of the actual sizes of the previous time periods' error terms; often the variance is related to the squares of the previous innovations. The ARCH model is appropriate when the error variance in a time series follows an autoregressive (AR) model; if an autoregressive moving average (ARMA) model is assumed for the error variance, the model is a generalized autoregressive conditional heteroskedasticity (GARCH) model.
ARCH models are commonly employed in modeling financial time series that exhibit time-varying volatility and volatility clustering, i.e. periods of swings interspersed with periods of relative calm. ARCH-type models are sometimes considered to be in the family of stochastic volatility models, although this is strictly incorrect since at time t the volatility is completely pre-determined (deterministic) given previous values.
To model a time series using an ARCH process, let denote the error terms (return residuals, with respect to a mean process), i.e. the series terms. These are split into a stochastic piece and a time-dependent standard deviation characterizing the typical size of the terms so that
The random variable is a strong white noise process. The series is modeled by
An ARCH(q) model can be estimated using ordinary least squares. A method for testing whether the residuals exhibit time-varying heteroskedasticity using the Lagrange multiplier test was proposed by Engle (1982). This procedure is as follows:
If an autoregressive moving average model (ARMA) model is assumed for the error variance, the model is a generalized autoregressive conditional heteroskedasticity (GARCH) model.
In that case, the GARCH (p, q) model (where p is the order of the GARCH terms and q is the order of the ARCH terms ), following the notation of the original paper, is given by
Generally, when testing for heteroskedasticity in econometric models, the best test is the White test. However, when dealing with time series data, this means to test for ARCH and GARCH errors.
Exponentially weighted moving average (EWMA) is an alternative model in a separate class of exponential smoothing models. As an alternative to GARCH modelling it has some attractive properties such as a greater weight upon more recent observations, but also drawbacks such as an arbitrary decay factor that introduces subjectivity into the estimation.
The lag length p of a GARCH(p, q) process is established in three steps:
Nonlinear Asymmetric GARCH(1,1) (NAGARCH) is a model with the specification:
For stock returns, parameter is usually estimated to be positive; in this case, it reflects a phenomenon commonly referred to as the "leverage effect", signifying that negative returns increase future volatility by a larger amount than positive returns of the same magnitude.
This model should not be confused with the NARCH model, together with the NGARCH extension, introduced by Higgins and Bera in 1992.
Integrated Generalized Autoregressive Conditional heteroskedasticity (IGARCH) is a restricted version of the GARCH model, where the persistent parameters sum up to one, and imports a unit root in the GARCH process. The condition for this is
The exponential generalized autoregressive conditional heteroskedastic (EGARCH) model by Nelson & Cao (1991) is another form of the GARCH model. Formally, an EGARCH(p,q):
where , is the conditional variance, , , , and are coefficients. may be a standard normal variable or come from a generalized error distribution. The formulation for allows the sign and the magnitude of to have separate effects on the volatility. This is particularly useful in an asset pricing context.
Since may be negative, there are no sign restrictions for the parameters.
The GARCH-in-mean (GARCH-M) model adds a heteroskedasticity term into the mean equation. It has the specification:
The residual is defined as:
The Quadratic GARCH (QGARCH) model by Sentana (1995) is used to model asymmetric effects of positive and negative shocks.
In the example of a GARCH(1,1) model, the residual process is
where is i.i.d. and
Similar to QGARCH, the Glosten-Jagannathan-Runkle GARCH (GJR-GARCH) model by Glosten, Jagannathan and Runkle (1993) also models asymmetry in the ARCH process. The suggestion is to model where is i.i.d., and
where if , and if .
The Threshold GARCH (TGARCH) model by Zakoian (1994) is similar to GJR GARCH. The specification is one on conditional standard deviation instead of conditional variance:
where if , and if . Likewise, if , and if .
Hentschel's fGARCH model, also known as Family GARCH, is an omnibus model that nests a variety of other popular symmetric and asymmetric GARCH models including APARCH, GJR, AVGARCH, NGARCH, etc.
In 2004, Claudia Klüppelberg, Alexander Lindner and Ross Maller proposed a continuous-time generalization of the discrete-time GARCH(1,1) process. The idea is to start with the GARCH(1,1) model equations
and then to replace the strong white noise process by the infinitesimal increments of a Lévy process , and the squared noise process by the increments , where
is the purely discontinuous part of the quadratic variation process of . The result is the following system of stochastic differential equations:
where the positive parameters , and are determined by , and . Now given some initial condition , the system above has a pathwise unique solution which is then called the continuous-time GARCH (COGARCH) model.
Unlike GARCH model, the Zero-Drift GARCH (ZD-GARCH) model by Li, Zhang, Zhu and Ling (2018)  lets the drift term in the first order GARCH model. The ZD-GARCH model is to model , where is i.i.d., and
The ZD-GARCH model does not require , and hence it nests the Exponentially weighted moving average (EWMA) model in "RiskMetrics". Since the drift term , the ZD-GARCH model is always non-stationary, and its statistical inference methods are quite different from those for the classical GARCH model. Based on the historical data, the parameters and can be estimated by the generalized QMLE method.
Spatial GARCH processes by Otto, Schmid and Garthoff (2018)  are considered as the spatial equivalent to the temporal generalized autoregressive conditional heteroscedasticity (GARCH) models. In contrast to the temporal ARCH model, in which the distribution is known given the full information set for the prior periods, the distribution is not straightforward in the spatial and spatiotemporal setting due to the interdependence between neighboring spatial locations. The spatial model is given by and
where denotes the -th spatial location and refers to the -th entry of a spatial weight matrix and for . The spatial weight matrix defines which locations are considered to be adjacent.
In a different vein, the machine learning community has proposed the use of Gaussian process regression models to obtain a GARCH scheme. This results in a nonparametric modelling scheme, which allows for: (i) advanced robustness to overfitting, since the model marginalises over its parameters to perform inference, under a Bayesian inference rationale; and (ii) capturing highly-nonlinear dependencies without increasing model complexity.
It is not yet clear in the finance literature that the asymmetric properties of variances are due to changing leverage. The name "leverage effect" is used simply because it is popular among researchers when referring to such a phenomenon.
Special attention to the model is given by the parameter of asymmetry [theta (θ)] which describes the correlation between returns and variance.6 ...
6 In the case of analyzing stock returns, the positive value of [theta] reflects the empirically well known leverage effect indicating that a downward movement in the price of a stock causes more of an increase in variance more than a same value downward movement in the price of a stock, meaning that returns and variance are negatively correlated