irf: VAR ipulse response functions

In gabrielrvsc/lbvar: Implementation Large Bayesian VAR

Description

Estimates impulse response coefficients of a lbvar model h steps ahead drawing from the posterior.

Usage

irf(object, ident, h, M = 100, unity.shock = TRUE)

Arguments

object	A lbvar object.
ident	A list with two elements: $A must be a matrix with the contemporaneus coefficients of a identified VAR and $sigma2u must be the structural shocks covariance matrix. This is the output of the function chol.identification.
h	Number of steps ahead.
M	Number of draws.
unity.shock	If TRUE the impulses are equal 1. If FALSE the impulses are of one standard deviation (default=TRUE).

Value

An object with S3 class "irf".

point.irf	A list with the point ir coefficients. Each element in the list is a matrix with the response on all variables cause by an impulse on the variable that gives name to the matrix.
density	A list that stores the ir coefficients from each draw. Each element in the list is another list with the response on all variables cause by an impulse on the variable that gives name to the list.

References

Garcia, Medeiros and Vasconcelos (2017).

See Also

predict, lbvar, identification, plot.irf

Examples

## == This example uses the Brazilian inflation data from
#Garcia, Medeiros and Vasconcelos (2017) == ##

# = This is an ilustrative example = #
# = The identification ignores which variables are more exogenous = #
data("BRinf")
Y=BRinf[,1:59]# remove expectation variables
modelB=lbvar(Y,p=3)
identB=identification(modelB)
irfB=irf(modelB,identB,h=12,M=100)
plot(irfB,1,2,alpha=0.1)

gabrielrvsc/lbvar documentation built on Aug. 22, 2021, 7:48 a.m.