Engle and Granger in 1987, recommended that «If a set of variables are cointegrated,
then there exists a valid error correction representation of the data, and
viceversa». In other words, this mean that if two variables are cointegrated
there must be some force that will return the equilibrium error back to zero.
Engle and Granger in 1987, also suggested a two-step model for cointegration
analysis. For example, let’s say that we have an independent variable
and a dependent one
First of all,
it should be estimated the long-run equilibrium equation:
We run a OLS
regression and we have:
We solve the
above equation with respect to
and we have:
In practice a
cointegration test is a test which testes if the residuals are stationary. To examine
this, we run a ADF test on the residuals, with the MacKinnon (1991) critical
values adjusted for the specific number of variables of our model. If the hypothesis
of the existence of cointegration cannot be rejected, the OLS estimator, is
said to be super-consistent. This means that for a very big T there is no need
variables in the model
The only thing that matters from the above test is the stationarity of
the residuals, if they are stationary we move to the second step. So, we save
the residuals and we prosed to the second step.
step we use the unit root process for the stationarity of the residuals to the
equation does not include constant term because the residuals have been
calculated with the method of ordinary lest squares, so they have zero mean.
The test suggested from the Engle Granger is a little bit different from those of
the one of Dickey-Fuller. The hypothesis of this test is:
The null hypothesis
can be rejected only when
(? is the critical value of Engle-Granger table).
Granger Test can be also used for more than two variables. The process is the
same with the one followed for two variables.
the cointegration process is a way to estimate the long run relation re else equilibrium
between two or even more variables. Engle and Granger in 1987 proved that if two
variables are cointegrated, then those variables have a long run relation
equilibrium, while in short run this may not be the case. The short run disequilibrium relation
can be proved with an Error Correction Mechanism (ECM). The equilibrium error
can be used to combine the long run with the short run with the help of ECM.
The equation of this model is:
: is the
: is the short
run coefficient which has to be less than zero and more than -1.
: are the
first differences of
We now can now
use the OLS because all the variables are
It is important
to point out that long run equilibrium is tested trough the significance of coincidence
is significant then
in the long run. In addition, the coefficient
measures the speed of adjustment toward the long
run equilibrium. The higher this coefficient the faster the return to the equilibrium.