Time-Varying Coefficient Arrays of the Lagged Endogenous Variables of a TVVAR (no intercept).

Source: R/tvA.R

Returns the estimated coefficients of the lagged endogenous variables as an array. Given an estimated time varying VAR of the form: $$\hat{\bold{y}}_t = \hat{A}_{1t} \bold{y}_{t-1} + \ldots + \hat{A}_{pt} \bold{y}_{t-p} + \hat{C}_tD_t$$ the function returns a list for each equation with \(\hat{A}_{1t} | \ldots | \hat{A}_{pt} | \hat{C}_t)\) set of arrays

tvAcoef(x)

Arguments

x

An object of class tvvar generated by tvVAR.

Value

A list object with coefficient arrays for the lagged endogenous variables.

Examples

data(Canada, package="vars")
var.2p <- vars::VAR(Canada, p = 2, type = "const")
tvvar.2p <- tvVAR(Canada, p = 2, type = "const")

#> Calculating regression bandwidths... bandwidth(s)  1.055559 4.118743 20 0.8864787 
A <- vars::Acoef(var.2p)
tvA <- tvAcoef(tvvar.2p)