Lecture Notes (4)

Modelling Volatility

Motivations: An Excursion into Non-linearity Land

• All of the models that have been discussed so far have been linear in
nature- that is, the model is linear in the parameters. The “traditional”
structural model could be something like:
yt =  1 +  2x2t + ... + kxkt + ut, or more compactly y = X + u.

• We also assumed ut  N(0, 2).

• Motivation: the linear structural (and time series) models cannot explain a
number of important features common to much financial data
– Leptokurtosis: the tendency for financial asset returns to have distributions that exhibit
fat tails and excess peakedness at the mean
– Volatility clustering or volatility pooling: the tendency for volatility in financial
markets to appear in bunches. Thus the large returns (of either sign) are expected to
follow large returns, and small returns (of either sign) to follow small returns.
– Leverage effects: the tendency for volatility to rise more following a large price fall
than following a price rise of the same magnitude.

2
 Non-linear Models: A Definition

• Campbell, Lo and MacKinlay (1997) define a non-linear data

generating process as one that can be written
yt = f(ut, ut-1, ut-2, …)
where ut is an iid error term and f is a non-linear function.

• They also give a slightly more specific definition as

yt = g(ut-1, ut-2, …)+ ut2(ut-1, ut-2, …)
where g is a function of past error terms only and 2 is a variance term.

• Models with nonlinear g(•) are “non-linear in mean”, while those with
nonlinear 2(•) are “non-linear in variance”. - ARCH / GARCH.

Models for Volatility

• Historical volatility models

• Implied volatility models

• Exponentially weighted moving average models

• Autoregressive volatility models

(Details refer to pages 420 ~423)

• Autoregressive conditionally heteroscedastic (ARCH) models

4
 Heteroscedasticity Revisited

• An example of a structural model is

with ut  N(0, s t2 ).

• The assumption that the variance of the errors is constant is known as

homoscedasticity, i.e. Var (ut) = s t2 .

• What if the variance of the errors is not constant?

- heteroscedasticity
- would imply that standard error estimates could be wrong.

• Is the variance of the errors likely to be constant over time? Not for
financial data.

A Sample Financial Asset Returns Time Series

Daily S&P 500 Returns for January 1990 – December 1999

6
 Autoregressive Conditionally Heteroscedastic
(ARCH) Models

• So use a model which does not assume that the variance is constant.
• Recall the definition of the variance of ut:
= Var(ut ut-1, ut-2,...) = E[(ut-E(ut))2 ut-1, ut-2,...]
We usually assume that E(ut) = 0
so = Var(ut  ut-1, ut-2,...) = E[ut2 ut-1, ut-2,...].

• What could the current value of the variance of the errors plausibly
depend upon?
– Previous squared error terms.
• This leads to the autoregressive conditionally heteroscedastic model
for the variance of the errors:
= 0 + 1
• This is known as an ARCH(1) model.

Autoregressive Conditionally Heteroscedastic

(ARCH) Models (cont’d)
• The full model would be
yt =  1 +  2x2t + ... + kxkt + ut , ut  N(0, )
where = 0 + 1
• We can easily extend this to the general case where the error variance
depends on q lags of squared errors:
= 0 + 1 +2 +...+q
• This is an ARCH(q) model.

• Instead of calling the variance , in the literature it is usually called ht,

so the model is
yt =  1 + 2x2t + ... + kxkt + ut , ut  N(0, ht)

where ht = 0 + 1 +2 +...+q

8
 Another Way of Writing ARCH Models

• For illustration, consider an ARCH(1). Instead of the above, we can

write

yt = 1 +  2x2t + ... + kxkt + ut , ut = vtt

, vt  N(0,1)

• The two are different ways of expressing exactly the same model. The
first form is easier to understand while the second form is required for
simulating from an ARCH model, for example.

Testing for “ARCH Effects”

1. First, run any postulated linear regression of the form given in the equation
above, e.g. yt =  1 + 2x2t + ... + kxkt + ut
saving the residuals, .

2. Then square the residuals, and regress them on q own lags to test for ARCH
of order q, i.e. run the regression

where vt is iid.
Obtain R2 from this regression

3. The test statistic is defined as TR2 (the number of observations multiplied

by the coefficient of multiple correlation) from the last regression, and is
distributed as a 2(q).

10
 Testing for “ARCH Effects” (cont’d)

4. The null and alternative hypotheses are

H0 : 1 = 0 and 2 = 0 and 3 = 0 and ... and q = 0
H1 : 1  0 or 2  0 or 3  0 or ... or q  0.

If the value of the test statistic is greater than the critical value from the
2 distribution, then reject the null hypothesis.

• Note that the ARCH test is also sometimes applied directly to returns
instead of the residuals from Stage 1 above.

11

Problems with ARCH(q) Models

• How do we decide on q?
• The required value of q might be very large
• Non-negativity constraints might be violated.
– When we estimate an ARCH model, we require i >0  i=1,2,...,q
(since variance cannot be negative)

• A natural extension of an ARCH(q) model which gets around some of

these problems is a GARCH model.

12
 Generalised ARCH (GARCH) Models

• Due to Bollerslev (1986). Allow the conditional variance to be dependent

upon previous own lags
• The variance equation is now

• This is a GARCH(1,1) model, which is like an ARMA(1,1) model for the

variance equation.
• We could also write

• Substituting into (1) for t-12 :

13

Generalised ARCH (GARCH) Models (cont’d)

• Now substituting into (2) for t-22

• An infinite number of successive substitutions would yield

• So the GARCH(1,1) model can be written as an infinite order ARCH model.

• We can again extend the GARCH(1,1) model to a GARCH(p,q):

14
 Generalised ARCH (GARCH) Models (cont’d)

• But in general a GARCH(1,1) model will be sufficient to capture the

volatility clustering in the data.

• Why is GARCH Better than ARCH?

- more parsimonious - avoids overfitting
- less likely to breech non-negativity constraints

15

The Unconditional Variance under the GARCH

Specification

• The unconditional variance of ut is given by

when

• is termed “non-stationarity” in variance

• is termed intergrated GARCH

• For non-stationarity in variance, the conditional variance forecasts will

not converge on their unconditional value as the horizon increases.

16
 Estimation of ARCH / GARCH Models

• Since the model is no longer of the usual linear form, we cannot use
OLS.

• We use another technique known as maximum likelihood.

• The method works by finding the most likely values of the parameters
given the actual data.

• More specifically, we form a log-likelihood function and maximise it.

17

Estimation of ARCH / GARCH Models (cont’d)

• The steps involved in actually estimating an ARCH or GARCH model

are as follows

1. Specify the appropriate equations for the mean and the variance - e.g. an
AR(1)- GARCH(1,1) model:

2. Specify the log-likelihood function to maximise:

3. The computer will maximise the function and give parameter values and
their standard errors

18
 Estimation of GARCH Models Using
Maximum Likelihood

• Now we have yt =  + yt-1 + ut , ut  N(0, )

• Unfortunately, the LLF for a model with time-varying variances cannot be

maximised analytically, except in the simplest of cases. So a numerical
procedure is used to maximise the log-likelihood function. A potential
problem: local optima or multimodalities in the likelihood surface.

• The way we do the optimisation is:

1. Set up LLF.
2. Use regression to get initial guesses for the mean parameters.
3. Choose some initial guesses for the conditional variance parameters.
4. Specify a convergence criterion - either by criterion or by value.

19

Non-Normality and Maximum Likelihood

• Recall that the conditional normality assumption for ut is essential.

• We can test for normality using the following representation

ut = vtt vt  N(0,1)

• The sample counterpart is

• Are the normal? Typically are still leptokurtic, although less so than
the . Is this a problem? Not really, as we can use the ML with a robust
variance/covariance estimator. ML with robust standard errors is called Quasi-
Maximum Likelihood or QML.

20
 Extensions to the Basic GARCH Model

• Since the GARCH model was developed, a huge number of extensions

and variants have been proposed. Three of the most important
examples are EGARCH, GJR, and GARCH-M models.

• Problems with GARCH(p,q) Models:

- Non-negativity constraints may still be violated
- GARCH models cannot account for leverage effects

• Possible solutions: the exponential GARCH (EGARCH) model or the

GJR model, which are asymmetric GARCH models.

21

The EGARCH Model

• Suggested by Nelson (1991). The variance equation is given by

• Advantages of the model

- Since we model the log(t2), then even if the parameters are negative, t2
will be positive.
- We can account for the leverage effect: if the relationship between
volatility and returns is negative, , will be negative.

22
 The GJR Model

• Due to Glosten, Jaganathan and Runkle

where It-1 = 1 if ut-1 < 0

= 0 otherwise

• For a leverage effect, we would see  > 0.

• We require 1 +   0 and 1  0 for non-negativity.

23

An Example of the use of a GJR Model

• Using monthly S&P 500 returns, December 1979- June 1998

• Estimating a GJR model, we obtain the following results.

24
 News Impact Curves

The news impact curve plots the next period volatility (ht) that would arise from various
positive and negative values of ut-1, given an estimated model.

News Impact Curves for S&P 500 Returns using Coefficients from GARCH and GJR
Model Estimates:

25

GARCH-in Mean

• We expect a risk to be compensated by a higher return. So why not let

the return of a security be partly determined by its risk?

• Engle, Lilien and Robins (1987) suggested the ARCH-M specification.

A GARCH-M model would be

•  can be interpreted as a sort of risk premium.

• It is possible to combine all or some of these models together to get

more complex “hybrid” models - e.g. an ARMA-EGARCH(1,1)-M
model.

26
 What Use Are GARCH-type Models?

• GARCH can model the volatility clustering effect since the conditional
variance is autoregressive. Such models can be used to forecast volatility.

• We could show that

Var (yt  yt-1, yt-2, ...) = Var (ut  ut-1, ut-2, ...)

• So modelling t2 will give us models and forecasts for yt as well.

• Variance forecasts are additive over time.

27

Forecasting Variances using GARCH Models

• Producing conditional variance forecasts from GARCH models uses a

very similar approach to producing forecasts from ARMA models.
• It is again an exercise in iterating with the conditional expectations
operator.
• Consider the following GARCH(1,1) model:
, ut  N(0,t2),
• What is needed is to generate are forecasts of T+12 T, T+22 T, ...,
T+s2 T where T denotes all information available up to and including
observation T.
• Adding one to each of the time subscripts of the above conditional
variance equation, and then two, and then three would yield the
following equations
T+12 = 0 + 1uT2 + T2
T+22 = 0 + 1 uT+12 +T+12
T+32 = 0 + 1 uT+22 +T+22

28
 Forecasting Variances
using GARCH Models (Cont’d)

• Let be the one step ahead forecast for 2 made at time T. This is
easy to calculate since, at time T, the values of all the terms on the
RHS are known.
• would be obtained by taking the conditional expectation of the
first equation:

• Given, how is , the 2-step ahead forecast for 2 made at time T,

calculated? Taking the conditional expectation of the second equation:
= 0 + 1E(  T) +
• where E(  T) is the expectation, made at time T, of , which is
the squared disturbance term.

29