fevd.id: Forecast Error Variance Decomposition

In pvars: VAR Modeling for Heterogeneous Panels

Forecast Error Variance Decomposition

Description

Calculates the forecast error variance decomposition. Respects SVAR models of cases S \neq K, i.e. partially identified or excess shocks, too.

Usage

## S3 method for class 'id'
fevd(x, n.ahead = 10, ...)

Arguments

x	SVAR object of class 'id' or any other that will be coerced to ('id', 'varx').
n.ahead	Integer specifying the steps ahead, i.e. the horizon of the FEVD.
...	Currently not used.

Value

A list of class 'svarfevd' holding the forecast error variance decomposition of each variables as a 'data.frame'.

References

Luetkepohl, H. (2005): New Introduction to Multiple Time Series Analysis, Springer, 2nd ed.

Jentsch, Lunsford (2022): "Asymptotically Valid Bootstrap Inference for Proxy SVARs", Journal of Business and Economic Statistics, 40, pp. 1876-1891.

Examples

data("PCIT")
names_k = c("APITR", "ACITR", "PITB", "CITB", "GOV", "RGDP", "DEBT")
names_l = c("m_PI", "m_CI")  # proxy names
names_s = paste0("epsilon[ ", c("PI", "CI"), " ]")  # shock names
dim_p   = 4  # lag-order

# estimate and identify proxy SVAR #
R.vars = vars::VAR(PCIT[ , names_k], p=dim_p, type="const")
R.idBL = id.iv(R.vars, iv=PCIT[-(1:dim_p), names_l], S2="MR", cov_u="OMEGA")
colnames(R.idBL$B) = names_s  # labeling

# calculate and plot FEVD under partial identification #
plot(fevd(R.idBL, n.ahead=20))

pvars documentation built on Nov. 5, 2025, 6:57 p.m.