R/plot.tscgm.ar2.r

In SparseTSCGM: Sparse Time Series Chain Graphical Models

#' Plot method for tscgm.ar2 objects
#'
#' Plots matrices from \code{tscgm.ar2} objects. You can choose to plot the precision matrix,
#' the first-order autoregressive matrix, or the second-order autoregressive matrix.
#'
#' @param x An object of class \code{tscgm.ar2}.
#' @param mat Character. Which matrix to plot. Choices are
#' \code{"precision"}, \code{"autoregression1"}, or \code{"autoregression2"}.
#' @param ... Additional plotting arguments passed to the plotting function.
#'
#' @export
#' @method plot tscgm.ar2
plot.tscgm.ar2 <- function(x, mat=c("precision","autoregression1", 
     "autoregression2"),...){
  mat=match.arg(mat)
  prec <- x$theta
  p=dim(prec)[2]
  if (mat == "precision") {
      prec <- prec
      nw_full <- network(prec)
      plot.network(nw_full,label = network.vertex.names(nw_full), 
         usearrows = FALSE,
         displayisolates = FALSE,...)
   }
 else if (mat == "autoregression1") {
      autoR1 <- x$gamma[1:p,] 
      nw_full <- network(autoR1, loops=TRUE)
      plot.network(nw_full,label = network.vertex.names(nw_full), 
        usearrows = TRUE,
        displayisolates = FALSE,...)
       }
else if (mat == "autoregression2") {
        d1= p + 1
        d2= 2* p
        autoR2 <- x$gamma[d1:d2,] 
        nw_full <- network(autoR2, loops=TRUE)
      plot.network(nw_full,label = network.vertex.names(nw_full), 
        usearrows = TRUE,
        displayisolates = FALSE, ...)
        }
}