PDFlib PLOP: PDF Linearization, Optimization, Protection

Page inserted by evaluation version

www.pdflib.com – sales@pdflib.com
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 67, 1 (2005) 0305-9049

Interpretation of Cointegrating Coefficients in

the Cointegrated Vector Autoregressive Model*
SØren Johansen

Department of Applied Mathematics and Statistics, University of Copenhagen,

Universitetsparken 5, 2100 Copenhagen Ø, Denmark (e-mail: sjo@math.ku.dk)

Abstract
Regression coefficients are interpreted by a counterfactual experiment. For
simultaneous equations this experiment can be implemented if the coefficients
are identified, and throws some light on the role of instruments and the method
of indirect least squares. This paper discusses another counterfactual
experiment in the vector autoregressive model in order to interpret the
coefficients of an identified cointegrating relation. The dynamics of the model
is used to implement a long-run change by changing the current values. The
counterfactual experiment can be conducted precisely when the cointegrating
relation is identified.

I. Introduction
The purpose of this note is to discuss the interpretation of cointegrating
coefficients as elasticities. Regression coefficients are usually interpreted by a
counterfactual experiment or thought experiment, which does not make
immediate sense in the cointegrated vector autoregressive (VAR) model,
because we do not have the dichotomy of endogenous and exogenous
variables.

*I would like to thank Giuseppe Bertola, Martin Browning, Tony Hall, Henrik Hansen, Jim
Heckman, David Hendry, Andrea Ichino, Alison Major, Adrian Pagan, and Tom Rothenberg for
useful comments. The support of the Danish Social Sciences Research Council and the ESF-EMM
network is gratefully acknowledged.
JEL Classification numbers: C32, C30.
93

Blackwell Publishing Ltd, 2005. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA.
94 Bulletin

The basic idea in this paper is to consider another thought experiment

which makes a distinction between what happens in the long run and the short
run, using the dynamics of the cointegrated model. Thus the dichotomy here is
that of current and long-run values. The note is inspired by Lütkepohl (1994),
who pointed out that there are problems related to the interpretation of
cointegrating coefficients as elasticities.
In section II, we discuss, using some examples, the well-known problem
of interpreting regression coefficients, and in section III we then turn to
cointegrating coefficients. In the Appendix, a few results on the companion
form of the cointegrated VAR model are collected.

II. The interpretation of regression coefficients

The idea of using counterfactuals as a basis for the concept of causality is of
course well known in statistics and econometrics. It was used by Haavelmo
(1944), Rubin (1974), Holland (1986) (see the survey by Angrist, Imbens and
Rubin, 1996 and the monograph by Hoover, 2001). We discuss here, the
interpretation of the coefficients in linear regression models, by considering
some simple examples.

Example 1
Consider first the univariate regression model
yt ¼ c1 x1t þ c2 x2t þ et
where we assume that the errors et are independent of the regressors xt. The
usual interpretation of the regression coefficient c1, say, is that c1 is the causal
effect of x1 on y, keeping x2 fixed. Thus, in order to interpret the coefficient c1
we conduct the thought experiment of changing the system by changing x1,
without changing x2, and observing the change in y. It is implicit in the
thought experiment that there is no other relation between y and x, and that we
can disregard relations (or correlation) between the xs and finally manipulate x
without changing e (independence between x and e).
Note that because
Eðyt jxt Þ ¼ c1 x1t þ c2 x2t
it follows that
@
Eðyt jxt Þ ¼ c1
@x1t
which is a convenient formulation of the role of c1 as a partial derivative
and the thought experiment above describes the definition of a partial
derivative.

Blackwell Publishing Ltd 2005

1 Interpretation of cointegrating coefficients 95

Example 2
As another example, we consider a system of simultaneous equations,
constructed to illustrate various situations
y1t
/2 y2t ¼ c11 x1t þ e1t
y2t
/3 y3t ¼ c22 x2t þ e2t ð1Þ
y2t
/4 y3t ¼ c31 x1t þ c32 x2t þ e3t :

In order to make sure that y is a function of x, and hence endogenous, we

assume that /3 6¼ /4, so that y can be solved as a function of x.
The first two equations are uniquely identified by the zero restrictions if the
parameters satisfy Wald’s condition

 

 /3 c22   
  6¼ 0  1 c11  6¼ 0
 / c   0 c 
4 32 31

while the third is not identified as there is only one restriction.

We interpret or give a meaning to the coefficient /2, which is a
coefficient to an endogenous variable, as the effect of y2 on y1 keeping x1
fixed.
According to Haavelmo (1944, p. 6), the equation and its coefficients are
defined by a ‘design of experiment’ to measure y1 as a function of y2 and x1,
by asking the question ‘what would be the value for y1 for various values of y2
and x1?’
Because the equation is here considered as part of a system which
simultaneously determines all ys as functions of xs, it is natural to consider
another experiment that only allows changes in x, before y is determined by
the equations. Thus, this second experiment is a way of implementing the
changes imagined in the first experiment. We therefore consider an experiment
where we keep x1 fixed and change x2 in order to produce the desired effect
which is 1 on y2 and /2 on y1. A change in x2 has no direct influence on y1,
because the coefficient on x2 is zero. Hence we use x2 as an instrument to
produce an effect in y2 through the second and third equation, which in turn
will help the interpretation of /2.
If we fix x1 and change x2 to x2 + h, then y is shifted to a new position, yh
satisfying
y1h
/2 y2h ¼ c11 x1
y2h
/3 y3h ¼ c22 ðx2 þ hÞ
y2h
/4 y3h ¼ c31 x1 þ c32 ðx2 þ hÞ:

Comparing the new position yh to y, we find

Blackwell Publishing Ltd 2005

 96 Bulletin

y1h
y1
/2 ðy2h
y2 Þ ¼ 0
y2h
y2
/3 ðy3h
y3 Þ ¼ c22 h
y2h
y2
/4 ðy3h
y3 Þ ¼ c32 h:
The solution is
y1h
y1 ¼ /2 ðy2h
y2 Þ
hð/4 c22
/3 c32 Þ
y2h
y2 ¼
ð/4
/3 Þ
hðc22
c32 Þ
y3h
y3 ¼ :
ð/4
/3 Þ
Hence, by choosing an increment of h ¼ (/4 ) /3)/(/4c22 ) /3c32) to x2 we find
that the effect on y2 is y2h
y2 ¼ 1 and that on y1 is y1h
y1 ¼ /2 , as required.
Thus we can conduct the thought experiment of using the instrument x2 to
produce a change of 1 in y2, disregarding the change in y3, thereby producing
the required change of /2 in y1. Note the role of the exogenous variable x2 as
instrument to produce the required effect, and in a sense implement the
2 ‘design of experiment’ of Haavelmo (1944).
The idea of influencing y1 via y2, using x2 as an instrument, and first
discussing the effect of x2 on y2 (regressing y2 on x2), and then using the
effect h(/4c22 ) /3c32)/(/4 ) /3) (fitted value ^y2 ) to measure the effect on y1
(regression of y1 on ^y2 Þ is clearly the idea behind the two-stage least squares
(2SLS) by Theil (1953a,b, 1971), although in this particular just-identified
case, it is just indirect least squares, which was used by Tinbergen (1930).
Note also that the condition for identification of the first equation is that
/4c22 ) c32/3 6¼ 0. Thus, in the construction of the thought experiment, we
apply the assumption that the three equations determine the endogenous
variables as functions of the exogenous (/3 6¼ /4), and the assumption that
the first equation is identified (/4c22 ) c32/3 6¼ 0).
Another way of expressing the interpretation of /2, is to note that from
equation (1) it follows that
Eðy1t jx1t ; x2t Þ ¼ /2 Eðy2t jx1t ; x2t Þ þ c11 x1t :
Here E(y1t|x1t, x2t) is considered a function of (x1t, x2t) through E(y2t|x1t, x2t)
and x1t, so that
@Eðy1t jx1t ; x2t Þ
/2 ¼ :
@Eðy2t jx1t ; x2t Þ
The counterfactual experiment then shows how one could, in principle,
implement the change in E(y2t|x1t, x2t) keeping x1t fixed, using x2t as an
instrument. Note that if x2t had a non-zero coefficient in the equation for y1t,

Blackwell Publishing Ltd 2005

1 Interpretation of cointegrating coefficients 97

the expectation would be a function of x1t and x2t as well as of E(y2t|x1t, x2t),
and the counterfactual experiment could not be performed, as one cannot keep
x1t and x2t fixed and change E(y2t|x1t, x2t). In this case, however, the parameter
/2 would not be uniquely identified in the system, although it would still have
a meaning according to the original ‘design of experiment’.

III. The interpretation of cointegrating coefficients

For the rest of the paper, we consider the cointegrated VAR model in n
dimensions with i.i.d. errors as given by the equation
X
k
1
Dxt ¼ ab0 xt
1 þ Ci Dxt
i þ et ; t ¼ 1; . . . ; T ð2Þ
i¼1

where we have left out deterministic terms to simplify the notation. Here, a
and b are n · r matrices of full rank. We assume that the process is I(1) so that

Xk
1  
 0 
a In
C b  6¼ 0
 ? i ? 
i¼1

(see Johansen, 1996, Theorem 4.2). For

X
k
1
C ¼ In
Ci
i¼1

we define
C ¼ b? ða0? Cb? Þ
1 a0?
and find from the solution to the equilibrium correction model in the
companion form [see equation (8) in the Appendix] the long-run value x1|t as
a function of current values (xt, . . . , xt)k+1) as given by
!
X
k
1
x1jt ¼ lim Eðxtþh jxt ; . . . ; xt
kþ1 Þ ¼ C xt
Ci xt
i : ð3Þ
h!1
i¼1

We first prove a simple result which defines two different thought

experiments.
Proposition 1. By adding h to all values xt)i, i ¼ 0, . . . , k)1, the long-run value
is changed by CCh 2 sp(b’). By adding h to xt, the long-run value is changed by
Ch 2 sp(b’). Conversely, a given long-run change k 2 sp(b’) can be achieved
by either adding k to all current values or by adding Ck to xt.
Proof. It is seen from equation (3) and the definition of C, that the effect of the
two types of changes are as indicated and proportional to b’.

Blackwell Publishing Ltd 2005

98 Bulletin

If, however, k 2 sp(b’) is the desired effect in the long run, then we write
k ¼ b’w, for some w 2 Rn)r, and find that a simultaneous change to all
current values by k will lead to the long-run change

CCk ¼ b? ða0? Cb? Þ
1 a0? Cb? w ¼ b? w ¼ k:

If we only change xt by adding Ck, we find the change CCk ¼ k. n

In the cointegrated VAR model, there are no exogenous and endogenous

variables and there is no natural way in which we can pick out some variables
and interpret the others as functions thereof. If we, nevertheless, solve the
stationary relations
b0 xt ¼ b01 x1t þ b02 x2t ¼ ut
for x1t, which we can do provided the r · r matrix b1 has full rank, or
equivalently that x2t does not cointegrate, then we find
x1t ¼ c0 x2t þ vt
where vt ¼ b0
1 0 0
1 0
1 ut is stationary and c ¼
b1 b2 . This is the usual regression
formulation, and the idea behind the regression estimator introduced by Engle
and Granger (1987), and the triangular representation of Phillips (1991). The
regressor x2t, however, is not independent of vt, and the usual interpretation
from section II has to be modified slightly. For large t, the correlation between
xt and vt is small and the process is approximately given by a random walk:
X
t
xt  C ei ;
i¼1

which again is approximately Nn(0, tR) with the singular long-run covariance
matrix R ¼ CXC¢. Because b¢C ¼ 0, we find from

0

0
b1 R11 R12 0
¼
b2 R21 R22 0
that
b01 R12 þ b02 R22 ¼ 0:

If x2t does not cointegrate, then R22 has full rank and we find

c0 ¼
b
10 0
1
1 b2 ¼ R12 R22

and hence the coefficient c¢ has the interpretation as a regression coefficient

derived from the singular long-run covariance matrix R.
There is, however, a dynamic structure in VAR models, which makes a
distinction between short- and long-run values or between current and

Blackwell Publishing Ltd 2005

1 Interpretation of cointegrating coefficients 99

long-run values (see Proposition 1). We apply this to consider a different type
of thought experiment, and use current values to implement a change in long-
run values. This thought experiment gives the possibility of implementing
changes in the long run needed for interpretation of the coefficients instead of,
as before, using a change in exogenous variables to produce a change in
endogenous variables. We first consider two examples and then give a general
result.

Example 3
Suppose the dimension n ¼ 3, and
b0 xt ¼ b1 x1t þ b2 x2t þ b3 x3t ¼ ut
is the only cointegrating relation in model (2), so that ut is stationary, whereas,
in general, the components xit are non-stationary. If b1 6¼ 0, we identify the
coefficients by solving for x1t, and find
x1t ¼ c2 x2t þ c3 x3t þ vt ð4Þ
with ci ¼ )bi/b1. We would like to interpret the coefficient c2 as the long-run
effect of x2 on x1, keeping x3 fixed; i.e. we would like to change x2 by 1, and x3
by 0, and observe a long-run change (c2) in x1. Because the relation is a long-
run relation, i.e. can be formulated as b¢x1|t ¼ 0, it is natural to think of these
changes as those in the long-run values and the idea is to accomplish these
changes by a suitable change in the current values, as formulated in
Proposition 1.
We therefore introduce a change to xt so that in the long run x2 is shifted by
1 and x3 is kept constant, while x1 is shifted by c2; i.e. we want to change
(x1, x2, x3)1|t to (x1, x2, x3)1|t + (c2, 1, 0). Now notice that the vector k ¼
(c2, 1, 0)¢ is orthogonal to b ¼ (1, )c2, )c3)¢:
k 0 b ¼ ðc2 ; 1; 0Þb ¼ c2
c2 ¼ 0
which means that the desired long-run change is a vector in the space spanned
by the columns of C, i.e. sp(b’). In general, we can ask the question if it is
possible to find a vector h and shift xt to xt + h so that the long-run value is
shifted from x1|t to x1|t + k, for a given vector k in sp(b’). This is exactly the
content of Proposition 1, where the relation between initial change and long-
run change is given, and it is seen that we can choose h ¼ Ck, and shift xt to
xt + Ck.
We use this result to interpret the coefficient c2 in the identified
cointegrating relation (4) as the effect of x2 on x1, in the sense that a
change of xt by C(c2, 1, 0)¢ will lead to a long-run change of the system
where x11|t is changed to x11|t + c2, x21|t to x21|t + 1, and x31|t is kept

Blackwell Publishing Ltd 2005

 100 Bulletin

fixed, as desired. In this sense, the coefficients of a cointegrating relation

can be considered long-run elasticities, provided the variables are measured
in logarithms.

Example 4
Consider next a cointegrated system with four variables and two cointegrating
relations, (b1, b2), which we identify by solving them for x1 and x2
x1t ¼ c13 x3t þ c14 x4t þ u1t
x2t ¼ c23 x3t þ c24 x4t þ u2t :
We want to implement a long-run change in the system, so that we can
interpret c13 as the long-run effect of x3 on x1 keeping x4 fixed. Because x2
does not enter the first relation, we consider a long-run change of the form
k 0 ðkÞ ¼ ðc13 ; k; 1; 0Þ
for some k, where the value of k is at our disposal, and we choose it so that
k(k) is orthogonal to b1 and b2. Note that
k 0 ðkÞb1 ¼ ðc13 ; k; 1; 0Þð1; 0;
c13 ;
c14 Þ0 ¼ 0
k 0 ðkÞb2 ¼ ðc13 ; k; 1; 0Þð0; 1;
c23 ;
c24 Þ0 ¼ k
c23 :
Thus, by the choice k ¼ c23 we make k(c23) orthogonal to both cointegrat-
ing vectors. Hence we can achieve the long-run change k ¼ k(c23) by
3 giving a change to xt of the form Ck, which by Proposition 1 will give a
long-run change of CCk ¼ k. Note that the long-run change also changes x2,
and it is the change of x2 that achieves that k(k) is orthogonal to both b
vectors. Thus, the role of the excluded variable x2 is that of an
instrument needed to produce the required effect on other variables
namely that the point stays on the attractor set, where the equations are
satisfied.
We finally show in Proposition 2 the general result that the identifying
restrictions define the instruments, that allow the interpretation of an identified
coefficient in a cointegrating relation. We construct a thought experiment that
achieves the required counterfactual change in the long-run value by suitable
changes in the current values.
We consider the model (2) for the cointegrated I(1) process Xt with r
cointegrating vectors. Let b1 be one of them of the form
b01 ¼ ð1; b21 ; . . . ; bs1 ; 0; . . . ; 0Þ
that is normalized on b11 ¼ 1, and restricted by R¢b1 ¼ 0, where
R0 ¼ ð0ðn
sÞs ; In
s Þ:

Blackwell Publishing Ltd 2005

1 Interpretation of cointegrating coefficients 101

We remind that the coefficient b21 is identified by the restrictions R, if for

any other vector b1 2 sp(b) satisfying b11 ¼ 1 and R0 b1 ¼ 0, it holds that
b21 ¼ b21 .

Proposition 2. Let b1 be a cointegrating vector of the form

b01 ¼ ð1; b21 ; . . . ; bs1 ; 0; . . . ; 0Þ:
The counterfactual experiment for implementing the definition of )b21 as the
long-run effect of x2 on x1 keeping x3, . . . , xs fixed and using xs+1, . . . , xn as
instruments, can be performed using a change in the current values xt, if and
only if b21 is identified by the zero restrictions.
Proof. Assume first that b21 is identified. We apply Proposition 1 to find the shift
needed to the current values. In order to apply this result, we construct a vector k
of the form
k 0 ðkÞ ¼ ð
b21 ; 1; 0; . . . ; 0; ksþ1 ; . . . ; kn Þ ¼ kð0Þ0 þ k0 R0 :
This vector is, by construction, orthogonal to b1 and we choose k¢ ¼ (ks+1, . . . ,
kn) so that it becomes orthogonal to all cointegrating vectors and becomes a
vector in sp(b’), in which case it can be implemented as a long-run change by
Proposition 1. Thus, we find k, so that
kðkÞ0 ðb2 ; . . . ; br Þ ¼ kð0Þ0 ðb2 ; . . . ; br Þ þ k0 R0 ðb2 ; . . . ; br Þ ¼ 0 : ð5Þ
This equation can be solved for k if the row vector k(0)¢(b2, . . . , br) is
spanned by the rows of R¢(b2, . . . , br), or equivalently if for any v so that
R¢(b2, . . . , br)v ¼ 0, it holds that k(0)¢(b2, . . . , br)v ¼ 0. Let v be a vector
such that R¢(b2, . . . , br)v ¼ 0, and define the vectors

b~1 ¼ ðb2 ; . . . ; br Þv
and
b1 ¼ ð1
b~11 Þb1 þ b~1 :

We now prove that kð0Þ0 b~1 ¼ kð0Þ0 ðb2 ; . . . ; br Þv ¼ 0:

Note that R0 b1 ¼ 0 because R¢b1 ¼ 0, and R0 b~1 ¼ 0; and that b11 ¼ 1;
because b11 ¼ 1. Therefore, because b21 is identified, we have b12 ¼ b12 and
b12 ¼ b12 ¼ ð1
b~11 Þb12 þ b~12
which implies that
b~11 b12 ¼ b~12 :
Hence
kð0Þ0 ðb2 ; . . . ; br Þv ¼ kð0Þ0 b~1 ¼
b12 b~11 þ b~12 ¼ 0

Blackwell Publishing Ltd 2005

102 Bulletin

as was to be shown. Thus, if b21 is identified, we can find a vector k which

solves equation (5) and hence Proposition 1 can be applied to implement the
desired long-run change.
If, however, b21 is not identified we can find a vector b1 ¼
ab1 þ ðb2 ; . . . ; br Þv for some a 2 R and v 2 Rr)1, with the property that
b11 ¼ 1; R0 b1 ¼ 0, and b21 6¼ b21 . We see that R0 ðb2 ; . . . ; br Þv ¼
R0 ðb1
ab1 Þ ¼ 0, and
kð0Þ0 ðb2 ; . . . ; br Þv ¼ kð0Þ0 ðb1
ab1 Þ
¼
b21 ð1
aÞ þ ðb21
ab21 Þ ¼ b21
b21 6¼ 0
and hence there cannot be a solution k to the equation (5). Thus the existence
of a k that solves equation (5) is necessary and sufficient for the identification
of b21, and hence for the construction of the counterfactual experiment.
Obviously a similar result can be proved for the interpretation of identified
regression coefficients in simultaneous equations, but we give it here only for
the case of cointegration.

IV. Conclusion
We have discussed interpretation of regression coefficients in simultaneous
equations and in coefficients in cointegrating relations. Whereas interpretation
of individual (autonomous) equations or cointegrating relations is in terms of a
counterfactual experiment that allows one to vary some variables and
keep others fixed, we discuss here how to implement such a change in terms
of another counterfactual experiment, when the equations form part of a
system of simultaneous equations. This experiment consists, for regression
coefficients, of manipulating the exogenous variables to achieve the relevant
effect on the endogenous variables. For cointegrating coefficients we
manipulate the current values in order to implement the required effect on
the long-run values. This last counterfactual experiment can be conducted
precisely when the coefficient is identified.

Final Manuscript Received: July 2004

References
Angrist, J. D., Imbens, G. W. and Rubin, D. B. (1996). ‘Identification of causal effects using
instrumental variables’, Journal of American Statistical Association, Vol. 91, pp. 444–472.
Engle, R. F. and Granger, C. W. J. (1987). ‘Co-integration and error correction: representation,
estimation and testing’, Econometrica, Vol. 55, pp. 251–276.
Haavelmo, T. (1944). ‘The probability approach in econometrics’, Econometrica, Vol. 12, Issue
Supplement, pp. 1–115.

Blackwell Publishing Ltd 2005

1 Interpretation of cointegrating coefficients 103

Hansen, P. and Johansen, S. (1998). Workbook on Cointegration, Oxford University Press,

Oxford.
Holland, P. (1986). ‘Statistics and causal inference’, Journal of American Statistical Associ-
ation, Vol. 81, pp. 945–970.
Hoover, K. D. (2001). Causality in Macroeconomics, Cambridge University Press, Cambridge.
Johansen, S. (1996). Likelihood-based Inference in Cointegrated Vector Autoregressive Models,
Oxford University Press, Oxford.
Lütkepohl, H. (1994). ‘Interpretation of cointegrating relations’, Econometric Reviews, Vol. 13,
pp. 391–394.
Phillips, P. C. B. (1991). ‘Optimal inference in cointegrated systems’, Econometrica, Vol. 59,
pp. 283–306.
Rubin, D. B. (1974). ‘Estimating causal effects of treatment in randomized and non randomized
studies’, Journal of Educational Psychology, Vol. 66, pp. 688–701.
Theil, H. (1953a). ‘Repeated least squares applied to complete equations systems’, Central
Planning Bureau, The Hague (mimeographed).
Theil, H. (1953b). ‘Estimation and simultaneous correlation in complete equation systems’,
Central Planning Bureau, The Hague (mimeographed).
Theil, H. (1971). Principles of Econometrics, North-Holland, Amsterdam.
Tinbergen, J. (1930). ‘Determination and interpretation of supply curves: An example’ [Bes-
timmung und Deutung von Angebotskurven. Ein Beispiel]. Zeitschrift für Nationalökonomie,
Vol. 1, pp. 779–679. Reprinted in Hendry D. F. and Morgan M. S. (1995). The Foundations
of Econometric Analysis, Cambridge University Press, Cambridge.

Appendix: The companion form

Model (2) can be expressed as an AR(1) model for the stacked process (see
Hansen and Johansen, 1998, exercise 4.7)

~xt ¼ ðx0t ; x0t
1 ; . . . ; x0t
kþ1 Þ0 ð6Þ

with errors
~et ¼ ðe0t ; 0; . . . ; 0Þ0

and parameters
0 1 0 1
a C1  Ck
1 b In  0
B0 In  0 C B 0
In  0 C
B C B C
a ¼ B ..
~ .. .. C; b~ ¼ B .. .. .. C ð7Þ
@. . . A @. . . A
0 0  In 0 0 
In
so that
0
ab~ ~xt
1 þ ~et
D~xt ¼ ~
with solution

Blackwell Publishing Ltd 2005

104 Bulletin

0 X
h
1
0
ab~ Þh~xt þ
~xtþh ¼ ðIkn þ ~ ðIkn þ ~ab~ Þi~etþh
i : ð8Þ
i¼0

Hence
0
Eð~xtþh j~xt Þ ¼ ðIkn þ ~ab~ Þh~xt : ð9Þ
The I(1) condition (Johansen, 1996, Theorem 4.2) is equivalent to
0
jeigðIrþðk
1Þn þ b~ ~aÞj < 1
0 h
which implies that ðIrþðk
1Þn þ b~ ~
aÞ ! 0. Applying this we find, using the
well-known relations
0 0 0
ab~ Þh ~
ðIkn þ ~ ab~ Þh
1 ðIkn þ ~ab~ Þ~a
a ¼ ðIkn þ ~
0 0
ab~ Þh
1 ~aðIrþðk
1Þn þ b~ ~aÞ
¼ ðIkn þ ~
0
aðIrþðk
1Þn þ b~ ~aÞh
¼  ¼ ~
and
0 0 0
ab~ Þh b~? ¼ ðIkn þ ~
ðIkn þ ~ ab~ Þh
1 ðIkn þ ~ab~ Þb~?
0
ab~ Þh
1 b~? ¼    ¼ b~?
¼ ðIkn þ ~
that
0 0 0 0
ab~ Þh~xt ¼ ðIkn þ ~
ðIkn þ ~ a0? b~? Þ
1 ~a0? þ ~aðb~ ~aÞ
1 b~ ~xt
ab~ Þh ½b~? ð~
0 0 0
a0? b~? Þ
1 ~
¼ b~? ð~ a0?~xt þ ~aðIrþðk
1Þn þ b~ ~aÞh ðb~ ~aÞ
1 b~ ~xt
! b~ ð~a0 b ~ Þ
1 ~a0 ~xt ¼ C~~ xt ¼ ~x1jt :
? ? ? ?

The limit ~x1jt is called the long-run value of ~xtþh starting at ~xt .
It is seen that
0
a0? ¼ a0? ðIn ;
C1 ; . . . ;
Ck
1 Þ;
~ b~? ¼ b0? ðIn ; . . . ; In Þ
which shows that
 
1
C~ ¼ b~? ~
a0? b~? a0? ¼ ðIn ; . . . ; In Þ0 CðIn ;
C1 ; . . . ;
Ck
1 Þ :
~ ð10Þ

For J ¢ ¼ (In,0, . . . , 0), we recover the process xt by xt ¼ J 0~xt, so that the

long-run value of xt+h starting at the current values (xt, . . . , xt)k+1) is
!
X
k
1
x1jt ¼ C xt
Ci xt
i :
i¼1

Blackwell Publishing Ltd 2005