Returns the estimated coefficient matrices of the moving average representation of a stable VAR(p), of an SVAR as an array or a converted VECM to VAR.
# S3 method for varest
Phi(x, nstep=10, ...)
# S3 method for svarest
Phi(x, nstep=10, ...)
# S3 method for svecest
Phi(x, nstep=10, ...)
# S3 method for vec2var
Phi(x, nstep=10, ...)
An array with dimension \((K \times K \times nstep + 1)\) holding the estimated coefficients of the moving average representation.
An object of class ‘varest
’, generated by
VAR()
, or an object of class ‘svarest
’,
generated by SVAR()
, or an object of class
‘svecest
’, generated by SVEC()
, or an object
of class ‘vec2var
’, generated by vec2var()
.
An integer specifying the number of moving error coefficient matrices to be calculated.
Currently not used.
Bernhard Pfaff
If the process \(\bold{y}_t\) is stationary (i.e. \(I(0)\), it has a Wold moving average representation in the form of: $$ \bold{y}_t = \Phi_0 \bold{u}_t + \Phi_1 \bold{u}_{t-1} + \Phi \bold{u}_{t-2} + \ldots , $$ whith \(\Phi_0 = I_k\) and the matrices \(\Phi_s\) can be computed recursively according to:
$$ \Phi_s = \sum_{j=1}^s \Phi_{s-j} A_j \quad s = 1, 2, \ldots , $$
whereby \(A_j\) are set to zero for \(j > p\). The matrix elements
represent the impulse responses of the components of \(\bold{y}_t\)
with respect to the shocks \(\bold{u}_t\). More precisely, the
\((i, j)\)th element of the matrix \(\Phi_s\) mirrors the expected
response of \(y_{i, t+s}\) to a unit change of the variable
\(y_{jt}\).
In case of a SVAR, the impulse response matrices are given by:
$$
\Theta_i = \Phi_i A^{-1} B \quad .
$$
Albeit the fact, that the Wold decomposition does not exist for
nonstationary processes, it is however still possible to compute the
\(\Phi_i\) matrices likewise with integrated variables or for the
level version of a VECM. However, a convergence to zero of
\(\Phi_i\) as i tends to infinity is not ensured; hence some shocks
may have a permanent effect.
Hamilton, J. (1994), Time Series Analysis, Princeton University Press, Princeton.
Lütkepohl, H. (2006), New Introduction to Multiple Time Series Analysis, Springer, New York.
Psi
, VAR
, SVAR
,
vec2var
, SVEC
data(Canada)
var.2c <- VAR(Canada, p = 2, type = "const")
Phi(var.2c, nstep=4)
Run the code above in your browser using DataLab