bv_irf: Impulse response settings and identification

Description

Provides settings for the computation of impulse responses to bvar, irf.bvar or fevd.bvar. Allows setting the horizon for which impulse responses should be computed, whether or not forecast error variance decompositions (FEVDs) should be included as well as if and what kind of identification should be used. See the Details section for further information on identification. Identification can be achieved via Cholesky decomposition, sign restrictions (Rubio-Ramirez, Waggoner and Zha, 2010), and zero and sign restrictions (Arias, Rubio-Ramirez and Waggoner, 2018).

Usage

bv_irf(
  horizon = 12,
  fevd = FALSE,
  identification = TRUE,
  sign_restr = NULL,
  sign_lim = 1000
)

Value

Returns a named list of class bv_irf with options for bvar, irf.bvar or fevd.bvar.

Arguments

horizon: Integer scalar. The horizon for which impulse responses (and FEVDs) should be computed. Note that the first period corresponds to impacts i.e. contemporaneous effects.
fevd: Logical scalar. Whether or not forecast error variance decompositions should be calculated.
identification: Logical scalar. Whether or not the shocks used for calculating impulses should be identified. Defaults to TRUE, i.e. identification via Cholesky decomposition of the VCOV-matrix unless sign_restr is provided.
sign_restr: Elements inform about expected impacts of certain shocks. Can be either $1$ , $- 1$ or $0$ depending on whether a positive, a negative or no contemporaneous effect of a certain shock is expected. Elements set to $N A$ indicate that there are no particular expectations for the contemporaneous effects. The default value is NULL. Note that in order to be fully identified at least $M * (M - 1) / 2$ restrictions have to be set and a maximum of $M - j$ zero restrictions can be imposed on the $j$ 'th column.
sign_lim: Integer scalar. Maximum number of tries to find suitable matrices to for fitting sign or zero and sign restrictions.

Details

Identification can be performed via Cholesky decomposition, sign restrictions, or zero and sign restrictions. The algorithm for generating suitable sign restrictions follows Rubio-Ramirez, Waggoner and Zha (2010), while the one for zero and sign restrictions follows Arias, Rubio-Ramirez and Waggoner (2018). Note the possiblity of finding no suitable zero/sign restrictions.

References

Rubio-Ramirez, J. F. and Waggoner, D. F. and Zha, T. (2010) Structural Vector Autoregressions: Theory of Identification and Algorithms for Inference. The Review of Economic Studies, 77, 665-696, tools:::Rd_expr_doi("10.1111/j.1467-937X.2009.00578.x"). Arias, J.E. and Rubio-Ramirez, J. F. and Waggoner, D. F. (2018) Inference Based on Structural Vector Autoregressions Identifiied with Sign and Zero Restrictions: Theory and Applications. Econometrica, 86, 2, 685-720, tools:::Rd_expr_doi("10.3982/ECTA14468").

Examples

Run this code

# Set impulse responses to a horizon of 20 time periods and enable FEVD
# (Identification is performed via Cholesky decomposition)
bv_irf(horizon = 20, fevd = TRUE)

# Set up structural impulse responses using sign restrictions
signs <- matrix(c(1, NA, NA, -1, 1, -1, -1, 1, 1), nrow = 3)
bv_irf(sign_restr = signs)

# Set up structural impulse responses using zero and sign restrictions
zero_signs <- matrix(c(1, 0, NA, -1, 1, 0, -1, 1, 1), nrow = 3)
bv_irf(sign_restr = zero_signs)

# Prepare to estimate unidentified impulse responses
bv_irf(identification = FALSE)

Run the code above in your browser using DataLab

Data Engineering and BI courses are free this week!