Recursively generates VAR data from initial values, coefficients, and
residuals. Optimised for maximum performance and memory efficiency.
Usage
fgenerateVARdata(y, p, c, beta, residuals)
Arguments
- y
T x N matrix of initial/historical endogenous variables.
- p
Integer lag order of the VAR model.
- c
Integer indicator for intercept (1 if included, 0 otherwise).
- beta
Coefficient matrix: (Np+c) x N if c = 1, or Np x N if c = 0.
- residuals
(T-p) x N matrix of residuals/shocks to add.
Value
T x N matrix of generated VAR data. The first p rows are copied
from y; the remaining T-p rows are generated recursively.
Details
The recursion follows:
$$y_t = c + A_1 y_{t-1} + \cdots + A_p y_{t-p} + e_t$$
Examples
if (FALSE) { # \dontrun{
y_init <- matrix(rnorm(200), ncol = 2)
beta <- matrix(rnorm(10), 5, 2)
residuals <- matrix(rnorm(190), 95, 2)
y_new <- fgenerateVARdata(y_init, p = 2, c = 1, beta, residuals)
} # }
