ragt2ridges (version 0.3.4)

mutualInfoVAR1: Mutual information analysis of the VAR(1) model

Description

Evaluate, within the VAR(1) model, the mutual information between each variate at the current time point and those at a future time point.

Usage

mutualInfoVAR1(A, SigmaE, T)

Arguments

A

Matrix \(\mathbf{A}\) of autoregression parameters.

SigmaE

Covariance matrix of the errors (innovations).

T

Positive numeric of length one specifying the future time point with which the mutual informations are to be evaluated.

Value

Object of class numeric with elements corresponding to the mutual informations. The \(j\)-th element represents the mutual information of the \(j\)-th variate at the current time point with all variates at the T-th time point from now.

References

Miok, V., Wilting, S.M., Van Wieringen, W.N. (2017), ``Ridge estimation of the VAR(1) model and its time series chain graph from multivariate time-course omics data'', Biometrical Journal, 59(1), 172-191.

See Also

ridgeVAR1.

Examples

# NOT RUN {
# set dimensions
p <- 3
n <- 4
T <- 10

# set model parameters
SigmaE <- diag(p)/4
A <- matrix(c(-0.1, -0.3,  0.6, 
               0.5, -0.4,    0, 
               0.3, -0.5, -0.2), byrow=TRUE, ncol=3)

# generate data
Y <- dataVAR1(n, T, A, SigmaE)

# fit VAR(1) model
VAR1hat <- ridgeVAR1(Y, 1, 1)

# impulse response analysis
mutualInfoVAR1(VAR1hat$A, solve(symm(VAR1hat$P)), T=5)
# }