Simulate data from VAR model

Description

Simulates a timeseries using VAR parameters

Usage

1
2
simulateVAR(pars,  means = 0, lags = 1, Nt = 100, init, residuals = 0.1,
                 burnin)

Arguments

pars

A square matrix or a list of square matrices indicating the VAR parameters

means

A vector of means.

lags

The lags to which the 'pars' argument parameters correspond. If 'pars' is a list then this argument should be a vector indicating which lags are represented by each element of the 'pars' list.

Nt

Number of time points

init

Initial setup. Must be a matrix of the first lags with rows corresponding to time points and columns corresponding to variables (e.g., if only two lags are used then the matrix must have two rows indicating the first two times points.)

residuals

Standard deviation of the residuals or a residual covariance matrix

burnin

Initial simulations not returned. Defaults to min(round(Nt/2), 100).

Author(s)

Sacha Epskamp (mail@sachaepskamp.com), Marie K. Deserno (m.k.deserno@uva.nl) and Laura F. Bringmann (laura.bringmann@ppw.kuleuven.be)