graphVAR1: Graphs of the temporal (or contemporaneous) relations implied...

In ragt2ridges: Ridge Estimation of Vector Auto-Regressive (VAR) Processes

Description

Graphs of the (conditional independence, temporal, or contemporaneous) relations among the variates as implied by the parameters of the VAR(1) model.

Usage

graphVAR1(sparseA, sparseP, type="TSCG", side="left", prune=TRUE, 
          nNames=NULL, main=NULL, vertex.color.T0="lightcyan2", 
          vertex.color.T1="lightcyan2", vertex.frame.color="steelblue", 
          vertex.label.cex=-1, vertex.label.color.T0="black", 
          vertex.label.color.T1="black", vertex.label.font=1.5, 
          vertex.size=-1, edge.arrow.size=-1, edge.width=-1, ...)

Arguments

sparseA	A matrix \mathbf{A} of autoregression parameters, which is assumed to be sparse.
sparseP	Precision matrix \mathbf{Ω}_{\varepsilon} the error, which is assumed to be sparse.
type	A character indicating what should be plotted. If type="TSCG". the time series chain graph is plotted, while if type="Aonly" limits this graph to the temporal relations. If type="globalPC" or type="contempPC", the global or contemporaneous (respectively) partial correlation graph is plotted.
side	A character indicating whether the contemporaneous dependencies should be plotted on the left-hand (time t) or the right-hand (time t+1) side. Only active when type="TSCG".
prune	A logical indicating whether to remove covariates without any temporal (or contemporaneous) relations (as implied by sparseA and sparseP).
nNames	A character containing the covariate names to be written inside the nodes.
main	The character to be plotted as title above the graph.
vertex.color.T0	Color of nodes at time point t.
vertex.color.T1	Color of nodes at time point t+1. This is ignored when type="globalPC" or type="contempPC".
vertex.frame.color	Refer to plot.igraph.
vertex.label.cex	Refer to plot.igraph.
vertex.label.color.T0	Color of the node label at time point t.
vertex.label.color.T1	Color of the node label at time point t+1. Ignored when type="globalPC" or type="contempPC".
vertex.label.font	Refer to plot.igraph.
vertex.size	Refer to plot.igraph.
edge.arrow.size	Refer to plot.igraph.
edge.width	Refer to plot.igraph.
...	Other arguments to be passed on to plot.igraph.

Author(s)

Wessel N. van Wieringen <[e-mail]>

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

CIGofVAR1, sparsify, sparsifyVAR1, plot.igraph.

Examples

# specify VAR(1) model parameters
A <- matrix(c(-0.1, -0.3,    0,  0.5,    0,    0, 
                 0,    0, -0.4, -0.1, -0.3,    0,  
               0.5,    0,    0,    0,    0, -0.4, 
              -0.1, -0.3,    0,  0.5,    0,    0, 
                 0,    0, -0.4, -0.1, -0.3,    0, 
               0.5,    0,    0,    0,    0, -0.4), byrow=TRUE, ncol=6)
P <- matrix(c(   2,    0, -0.5,  0.5,    0,  0.5, 
                 0,    1,  0.5,  0.5,  0.5,    0, 
              -0.5,  0.5,    1,    0,    0,  0.5, 
               0.5,  0.5,    0,    1,    0,    0, 
                 0,  0.5,    0,    0,    1,    0, 
               0.5,    0,  0.5,    0,    0,    1), byrow=TRUE, ncol=6)

# adjacency matrix of (global) conditional independencies.
graphVAR1(A, P, type="TSCG")

ragt2ridges documentation built on Jan. 28, 2020, 5:08 p.m.