Bootstrap VAR Model

Generates a bootstrap pseudo-sample from a fitted VAR model using either residual (iid) resampling or a wild (Rademacher) bootstrap scheme.

Usage

fbootstrapVAR(y, fVAR_result, bootscheme = "residual")

Arguments

y

T x N matrix of original endogenous variables.

fVAR_result

List from fVAR containing beta, residuals, p, and c.

bootscheme

String indicating the bootstrap method: "residual" for iid resampling of residuals, or "wild" for a Rademacher wild bootstrap.

Value

A list with two elements: ynext — T x N bootstrapped data matrix (first p rows copied from y); rademacher — T_eff-length vector of Rademacher signs used in the wild scheme (all zeros for residual scheme).

Details

The recursion is $$y_t = c + A_1 y_{t-1} + \cdots + A_p y_{t-p} + e_t^*$$ where \(e_t^*\) is either a randomly resampled residual (residual scheme) or the t-th residual multiplied by a Rademacher sign (wild scheme).

See also

fbootstrapChol, fVAR