createA: Generation of the VAR(1) autoregression coefficient matrix.

Description

Generates autoregression coefficient matrices of the VAR(1) with various type of topologies

Usage

createA(p, topology, nonzeroA=0, nCliques=1, nHubs=1, 
        nBands=1, percZeros=0.9, stationary=TRUE)

Arguments

A positive integer specifying the dimension of the square matrix \(\mathbf{A}\).

topology

Topology to impose on \(\mathbf{A}\): a character equalling either clique, hub, chain, or random.

nonzeroA

Numeric, value that nonzero elements of \(\mathbf{A}\) will assume. If equal to zero, a random value from the interval [-1,1] is sampled.

nCliques

When topology="clique", this positive integer specifies number of cliques.

nHubs

When topology="hub", this positive integer specifies number of hubs.

nBands

When topology="chain", this positive integer specifies number of bands.

percZeros

When topology="random", the probability with which zero elements of \(\mathbf{A}\) are to be sampled.

stationary

A logical: should the generated \(\mathbf{A}\) be stationary?

Value

A matrix with autoregression coefficient matrix \(\mathbf{A}\) of the VAR(1) model.

References

Miok, V., Wilting, S.M., Van Wieringen, W.N. (2016), "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.

Examples

Run this code

# NOT RUN {
# create the VAR(1) parameters
A  <- createA(10, topology="clique", nonzeroA=0.1, nClique=4)
Se <- createS(1000, 10, "star")

# sample data from the VAR(1) model with above parameters
Y <- dataVAR1(4, 8, A, Se)
# }