R/plot.postcp.R

In malithj/postCP: A package to estimate posterior probabilities and perform model selection in change-point models using constrained HMM

plot.postcp <-
function(x, y=NULL, xlab = "index",
         ylab = "probability", main = "Posterior Change Point Probability Distribution", p.col = "blue", pch = 16, p.cex = NA, m.col = "brown", m.lty = 2,
         m.lwd = 1, l.col = NA, l.lty = NA, l.lwd = NA, ...)
{
  # x: results from function postCP
  if (missing(x)) {
    stop("Need to use results of postCP first")
  }
  n <- x$n

  #Create plot
  plot(c(1,n), c(0,max(x$post.cp)), t = "n", xlab = xlab, ylab = ylab, col = p.col, pch = pch,
       cex = p.cex)
  #Change points
  abline( v = x$cp, col =m.col, lty = m.lty, lwd = m.lwd);
  #Add points to the graph
  for (k in 1:ncol(x$post.cp))
    points(x$post.cp[,k], t = 'l', col = k, lty = k)
  title( main = main)

}