The pth-order VAR process is written as
with .
Equivalently, it can be written as
with .
For stationarity, the VAR process must be expressible in the convergent causal infinite MA form as
where with , where denotes a norm for the matrix such as . The matrix can be recursively obtained from the relation ; it is
where and for .
The stationarity condition is satisfied if all roots of are outside of the unit circle. The stationarity condition is equivalent to the condition in the corresponding VAR(1) representation, , that all eigenvalues of the companion matrix be less than one in absolute value, where , , and
If the stationarity condition is not satisfied, a nonstationary model (a differenced model or an error correction model) might be more appropriate.
The following statements estimate a VAR(1) model and use the ROOTS option to compute the characteristic polynomial roots:
proc varmax data=simul1; model y1 y2 / p=1 noint print=(roots); run;
Figure 35.44 shows the output associated with the ROOTS option, which indicates that the series is stationary since the modulus of the eigenvalue is less than one.
Consider the stationary VAR(p) model
where are assumed to be available (for convenience of notation). This can be represented by the general form of the multivariate linear model,
where
with vec denoting the column stacking operator.
The conditional least squares estimator of is
and the estimate of is
where is the residual vectors. Consistency and asymptotic normality of the LS estimator are that
where converges in probability to and denotes convergence in distribution.
The (conditional) maximum likelihood estimator in the VAR(p) model is equal to the (conditional) least squares estimator on the assumption of normality of the error vectors.
As before, vec denotes the column stacking operator and vech is the corresponding operator that stacks the elements on and below the diagonal. For any matrix , the commutation matrix is defined as ; the duplication matrix is defined as ; the elimination matrix is defined as .
The asymptotic distribution of the impulse response function (Lütkepohl, 1993) is
where and
where is a matrix and is a companion matrix.
The asymptotic distribution of the accumulated impulse response function is
where .
The asymptotic distribution of the orthogonalized impulse response function is
where , , ,
and with and .
Let be arranged and partitioned in subgroups and with dimensions and , respectively (); that is, with the corresponding white noise process . Consider the VAR(p) model with partitioned coefficients for as follows:
The variables are said to cause , but do not cause if . The implication of this model structure is that future values of the process are influenced only by its own past and not by the past of , where future values of are influenced by the past of both and . If the future are not influenced by the past values of , then it can be better to model separately from .
Consider testing , where C is a matrix of rank s and c is an s-dimensional vector where . Assuming that
you get the Wald statistic
For the Granger causality test, the matrix C consists of zeros or ones and c is the zero vector. See Lütkepohl (1993) for more details of the Granger causality test.
The vector autoregressive model with exogenous variables is called the VARX(p,s) model. The form of the VARX(p,s) model can be written as
The parameter estimates can be obtained by representing the general form of the multivariate linear model,
where
The conditional least squares estimator of can be obtained by using the same method in a VAR(p) modeling. If the multivariate linear model has different independent variables that correspond to dependent variables, the SUR (seemingly unrelated regression) method is used to improve the regression estimates.
The following example fits the ordinary regression model:
proc varmax data=one; model y1-y3 = x1-x5; run;
This is equivalent to the REG procedure in the SAS/STAT software:
proc reg data=one; model y1 = x1-x5; model y2 = x1-x5; model y3 = x1-x5; run;
The following example fits the second-order lagged regression model:
proc varmax data=two; model y1 y2 = x / xlag=2; run;
This is equivalent to the REG procedure in the SAS/STAT software:
data three; set two; xlag1 = lag1(x); xlag2 = lag2(x); run; proc reg data=three; model y1 = x xlag1 xlag2; model y2 = x xlag1 xlag2; run;
The following example fits the ordinary regression model with different regressors:
proc varmax data=one; model y1 = x1-x3, y2 = x2 x3; run;
This is equivalent to the following SYSLIN procedure statements:
proc syslin data=one vardef=df sur; endogenous y1 y2; model y1 = x1-x3; model y2 = x2 x3; run;
From the output in Figure 35.20 in the section Getting Started: VARMAX Procedure, you can see that the parameters, XL0_1_2, XL0_2_1, XL0_3_1, and XL0_3_2 associated with the exogenous variables, are not significant. The following example fits the VARX(1,0) model with different regressors:
proc varmax data=grunfeld; model y1 = x1, y2 = x2, y3 / p=1 print=(estimates); run;
As you can see in Figure 35.45, the symbol ‘_’ in the elements of matrix corresponds to endogenous variables that do not take the denoted exogenous variables.