plot.mc.irf {MSBVAR} | R Documentation |
Provides a plotting method for the mc.irf
Monte Carlo
sample of impulse responses. Responses can be plotted with classical
or Bayesian error bands, as suggested by Sims and Zha (1999).
plot.mc.irf(x, method=c("Sims-Zha2"), component=1, probs=c(0.16,0.84), varnames = NULL, ...)
x |
Output of the mc.irf function |
method |
Method to be used for the error band construction.
Default method is to use the eigendecomposition method proposed by
Sims and Zha. Defined methods are "Percentile" (error bands are
based on percentiles specified in probs ), "Normal
Approximation" (Gaussian approximation for interval of width
probs ), "Sims-Zha1" (Gaussian approximation with linear
eigendecomposition), "Sims-Zha2" (Percentiles with
eigendecomposition for each impulse response function), "Sims-Zha3"
(Percentiles with eigendecomposition of the full stacked impulse responses) |
component |
If using one of the eigendecomposition methods, the
eigenvector component to be used for the error band
construction. Default is the first or largest eigenvector component. |
probs |
probs is the width of the error bands. Default
is c(0.16, 0.84) which is a 68% band that is approximately
one standard deviation, as suggested by Sims and Zha. |
varnames |
List of variable names for labeling the impulse responses. |
... |
Other graphics parameters |
This function plots the output of a Monte Carlo simulation of VAR
impulse response functions produced by mc.irf
. The function
allows the user to choose among a variety of frequentist (normal
appproximation and percentile) and Bayesian (eigendecomposition)
methods for constructing error bands around a set of impulse
responses. Impulses or shocks are in the columns and the rows are the
responses.
The primary reason for this function is to plot impulse responses and their error bands. Secondarily, it returns an invisible list of the impulses responses, their error bands, and summary measures of the fractions of the variance in the eigenvector methods that explain the total variation of each response.
responses |
Responses and their error bands |
eigenvector.fractions |
Fraction of the variation in each
response that is explained by the chosen eigenvectors. NULL
for non-eigenvector methods. |
Patrick T. Brandt
Brandt, Patrick T. and John R. Freeman. 2006. "Advances in Bayesian Time Series Modeling and the Study of Politics: Theory Testing, Forecasting, and Policy Analysis" Political Analysis 14(1):1-36.
Sims, C.A. and Tao Zha. 1999. "Error Bands for Impulse Responses." Econometrica. 67(5): 1113-1156.
See Also mc.irf
for the computation of Monte
Carlo samples of impulse responses,
szbsvar
for estimation of the posterior
moments of the B-SVAR model,
gibbs.A0.BSVAR
for Gibbs sampling the
posterior of the A(0) for the model, and
## Not run: data(IsraelPalestineConflict) fit.BVAR <- szbvar(IsraelPalestineConflict, p=6, z=NULL, lambda0=0.6, lambda1=0.1, lambda3=2, lambda4=0.5, lambda5=0, mu5=0, mu6=0, nu=3, qm=4, prior=0, posterior.fit=FALSE) posterior.impulses <- mc.irf(fit.BVAR, nsteps=12, draws=1000) plot(posterior.impulses, method = c("Percentile")) ## End(Not run)