plot.bsts.predictors function

Plot the most likely predictors

Creates a time series plot showing the most likely predictors of a time series used to fit a bsts object.

PlotBstsPredictors(bsts.object,
                     burn = SuggestBurn(.1, bsts.object),
                     inclusion.threshold = .1,
                     ylim = NULL,
                     flip.signs = TRUE,
                     show.legend = TRUE,
                     grayscale = TRUE,
                     short.names = TRUE,
                     ...)

Arguments

• bsts.object: An object of class bsts.
• burn: The number of observations you wish to discard as burn-in.
• inclusion.threshold: Plot predictors with marginal inclusion probabilities above this threshold.
• ylim: Scale for the vertical axis.
• flip.signs: If true then a predictor with a negative sign will be flipped before being plotted, to better align visually with the target series.
• show.legend: Should a legend be shown indicating which predictors are plotted?
• grayscale: Logical. If TRUE then lines for different predictors grow progressively lighter as their inclusion probability decreases. If FALSE then lines are drawn in black.
• short.names: Logical. If TRUE then a common prefix or suffix shared by all the variables will be discarded.
• ...: Extra arguments to be passed to plot.

Examples

data(AirPassengers)
  y <- log(AirPassengers)
  lag.y <- c(NA, head(y, -1))
  ss <- AddLocalLinearTrend(list(), y)
  ss <- AddSeasonal(ss, y, nseasons = 12)
  ## Call bsts with na.action = na.omit to omit the leading NA in lag.y
  model <- bsts(y ~ lag.y, state.specification = ss, niter = 500,
                na.action = na.omit)
  plot(model, "predictors")

See Also

bsts

PlotDynamicDistribution

plot.lm.spike