Vector Error Correction Model

A vector error correction model (VECM) can lead to a better understanding of the nature of any nonstationarity among the different component series and can also improve longer term forecasting over an unconstrained model.

The VECM() form with the cointegration rank is written as

where is the differencing operator, such that ; , where and are matrices; is a matrix.

It has an equivalent VAR() representation as described in the preceding section.

where is a identity matrix.

Example of Vector Error Correction Model

An example of the second-order nonstationary vector autoregressive model is

with

This process can be given the following VECM(2) representation with the cointegration rank one:

The following PROC IML statements generate simulated data for the VECM(2) form specified above and plot the data as shown in Figure 32.12:

proc iml;
   sig = 100*i(2);
   phi = {-0.2 0.1, 0.5 0.2, 0.8 0.7, -0.4 0.6};
   call varmasim(y,phi) sigma=sig n=100 initial=0
                        seed=45876;
   cn = {'y1' 'y2'};
   create simul2 from y[colname=cn];
   append from y;
quit;

data simul2;
   set simul2;
   date = intnx( 'year', '01jan1900'd, _n_-1 );
   format date year4. ;
run;

proc timeseries data=simul2 vectorplot=series;
   id date interval=year;
   var y1 y2;
run;

Figure 32.12 Plot of Generated Data Process

Cointegration Testing

The following statements use the Johansen cointegration rank test. The COINTTEST=(JOHANSEN) option does the Johansen trace test and is equivalent to specifying COINTTEST with no additional options or the COINTTEST=(JOHANSEN=(TYPE=TRACE)) option.

/*--- Cointegration Test ---*/

proc varmax data=simul2;
   model y1 y2 / p=2 noint dftest cointtest=(johansen);
run;

Figure 32.13 shows the output for Dickey-Fuller tests for the nonstationarity of each series and Johansen cointegration rank test between series.

Figure 32.13 Dickey-Fuller Tests and Cointegration Rank Test

The VARMAX Procedure

Unit Root Test
Variable	Type	Rho	Pr < Rho	Tau	Pr < Tau
y1	Zero Mean	1.47	0.9628	1.65	0.9755
	Single Mean	-0.80	0.9016	-0.47	0.8916
	Trend	-10.88	0.3573	-2.20	0.4815
y2	Zero Mean	-0.05	0.6692	-0.03	0.6707
	Single Mean	-6.03	0.3358	-1.72	0.4204
	Trend	-50.49	0.0003	-4.92	0.0006

Cointegration Rank Test Using Trace
H0: Rank=r	H1: Rank>r	Eigenvalue	Trace	5% Critical Value	Drift in ECM	Drift in Process
0	0	0.5086	70.7279	12.21	NOINT	Constant
1	1	0.0111	1.0921	4.14

In Dickey-Fuller tests, the second column specifies three types of models, which are zero mean, single mean, or trend. The third column ( Rho ) and the fifth column ( Tau ) are the test statistics for unit root testing. Other columns are their -values. You can see that both series have unit roots. For a description of Dickey-Fuller tests, see the section PROBDF Function for Dickey-Fuller Tests in Chapter 5, SAS Macros and Functions.

In the cointegration rank test, the last two columns explain the drift in the model or process. Since the NOINT option is specified, the model is

The column Drift In ECM means there is no separate drift in the error correction model, and the column Drift In Process means the process has a constant drift before differencing.

H0 is the null hypothesis, and H1 is the alternative hypothesis. The first row tests against ; the second row tests against . The Trace test statistics in the fourth column are computed by where is the available number of observations and is the eigenvalue in the third column. By default, the critical values at 5% significance level are used for testing. You can compare the test statistics and critical values in each row. There is one cointegrated process in this example since the Trace statistic for testing against is greater than the critical value, but the Trace statistic for testing against is not greater than the critical value.

The following statements fit a VECM(2) form to the simulated data. From the result in Figure 32.13, the time series are cointegrated with rank=1. You specify the ECM= option with the RANK=1 option. For normalizing the value of the cointegrated vector, you specify the normalized variable with the NORMALIZE= option. The PRINT=(IARR) option provides the VAR(2) representation. The VARMAX procedure output is shown in Figure 32.14 through Figure 32.16.

/*--- Vector Error-Correction Model ---*/

proc varmax data=simul2;
   model y1 y2 / p=2 noint lagmax=3
                 ecm=(rank=1 normalize=y1)
                 print=(iarr estimates);
run;

The ECM= option produces the estimates of the long-run parameter, , and the adjustment coefficient, . In Figure 32.14, "1" indicates the first column of the and matrices. Since the cointegration rank is 1 in the bivariate system, and are two-dimensional vectors. The estimated cointegrating vector is . Therefore, the long-run relationship between and is . The first element of is 1 since is specified as the normalized variable.