Ans = get(V,Query)
[Ans,Ans,...] = get(V,Query,Query,...)
V
[ VAR ] - VAR object.
Query
[ char ] - Query to the VAR object.
Ans
[ ... ] - Answer to the query.'yList'
-- Returns [ cellstr ] the names of endogenous variables.
'eList'
-- Returns [ cellstr ] the names of residuals or shocks.
'iList'
-- Returns [ cellstr ] the names of conditioning (forecast) instruments.
'ny'
-- Returns [ numeric ] the number of variables.
'ne'
-- Returns [ numeric ] the number of residuals or shocks.
'ni'
-- Returns [ numeric ] the number of conditioning (forecast) instruments.
'A#'
, 'A*'
, 'A$'
-- Returns [ numeric ] the transition matrix in one of the three possible forms; see Description.
'K'
, 'const'
-- Returns [ numeric ] the constant vector or matrix (the latter for panel VARs).
'J'
-- Returns [ numeric ] the coefficient matrix in front of exogenous inputs.
'Omg'
, 'Omega'
-- Returns [ numeric ] the covariance matrix of one-step-ahead forecast errors, i.e. reduced-form residuals. Note that this query returns the same matrix also for structural VAR (SVAR) objects.
'Sgm'
, 'Sigma'
-- Returns [ numeric ] the covariance matrix of the VAR parameter estimates; the matrix is non-empty only if the option 'covParam='
has been set to true
at estimation time.
'G'
-- Returns [ numeric ] the coefficient matrix on cointegration terms.
'AIC'
-- Returns [ numeric ] Akaike information criterion.
'SBC'
-- Returns [ numeric ] Schwarz bayesian criterion.
'cumLong'
-- Returns [ numeric ] the matrix of long-run cumulative responses.
'nFree'
-- Returns [ numeric ] the number of freely estimated (hyper-) parameters.
'order'
, 'p'
-- Returns [ numeric ] the order of the VAR object.
There are three queries to request the VAR transition matrix: 'A#'
, 'A*'
, 'A$'
. They differ in how the higher-order transition matrices are arranged.
'A#'
returns cat(3,I,-A1,...,-Ap)
where I
is an identity matrix, and A1
, ... Ap
are the coefficient matrices on individual lags.
'A#'
returns cat(3,A1,...,Ap)
where A1
, ... Ap
are the coefficient matrices on individual lags.
'A$'
returns [A1,...,Ap]
where A1
, ... Ap
are the coefficient matrices on individual lags.