Computes posterior draws of impulse responses

Each of the draws from the posterior estimation of a model is transformed into a draw from the posterior distribution of the impulse responses.

Usage

compute_impulse_responses(posterior, horizon, standardise = FALSE)

Arguments

posterior: posterior estimation outcome - an object of either of the classes: PosteriorBSVAR, PosteriorBSVARMSH, PosteriorBSVARMIX, or PosteriorBSVARSV obtained by running the estimate function. The interpretation depends on the normalisation of the shocks using function normalise_posterior(). Verify if the default settings are appropriate.
horizon: a positive integer number denoting the forecast horizon for the impulse responses computations.
standardise: a logical value. If TRUE, the impulse responses are standardised so that the variables' own shocks at horizon 0 are equal to 1. Otherwise, the parameter estimates determine this magnitude.

Value

An object of class PosteriorIR, that is, an NxNx(horizon+1)xS array with attribute PosteriorIR containing S draws of the impulse responses.

References

Kilian, L., & Lütkepohl, H. (2017). Structural VAR Tools, Chapter 4, In: Structural vector autoregressive analysis. Cambridge University Press.

Author

Tomasz Woźniak wozniak.tom@pm.me

Examples

# upload data
data(us_fiscal_lsuw)

# specify the model and set seed
set.seed(123)
specification  = specify_bsvar$new(us_fiscal_lsuw, p = 1)
#> The identification is set to the default option of lower-triangular structural matrix.

# run the burn-in
burn_in        = estimate(specification, 10)
#> **************************************************|
#> bsvars: Bayesian Structural Vector Autoregressions|
#> **************************************************|
#>  Gibbs sampler for the SVAR model                 |
#> **************************************************|
#>  Progress of the MCMC simulation for 10 draws
#>     Every 10th draw is saved via MCMC thinning
#>  Press Esc to interrupt the computations
#> **************************************************|

# estimate the model
posterior      = estimate(burn_in, 50)
#> **************************************************|
#> bsvars: Bayesian Structural Vector Autoregressions|
#> **************************************************|
#>  Gibbs sampler for the SVAR model                 |
#> **************************************************|
#>  Progress of the MCMC simulation for 50 draws
#>     Every 10th draw is saved via MCMC thinning
#>  Press Esc to interrupt the computations
#> **************************************************|

# compute impulse responses 2 years ahead
irf           = compute_impulse_responses(posterior, horizon = 8)

# workflow with the pipe |>
############################################################
set.seed(123)
us_fiscal_lsuw |>
  specify_bsvar$new(p = 1) |>
  estimate(S = 10) |> 
  estimate(S = 50) |> 
  compute_impulse_responses(horizon = 8) -> ir
#> The identification is set to the default option of lower-triangular structural matrix.
#> **************************************************|
#> bsvars: Bayesian Structural Vector Autoregressions|
#> **************************************************|
#>  Gibbs sampler for the SVAR model                 |
#> **************************************************|
#>  Progress of the MCMC simulation for 10 draws
#>     Every 10th draw is saved via MCMC thinning
#>  Press Esc to interrupt the computations
#> **************************************************|
#> **************************************************|
#> bsvars: Bayesian Structural Vector Autoregressions|
#> **************************************************|
#>  Gibbs sampler for the SVAR model                 |
#> **************************************************|
#>  Progress of the MCMC simulation for 50 draws
#>     Every 10th draw is saved via MCMC thinning
#>  Press Esc to interrupt the computations
#> **************************************************|