Plot forecast uncertainty contributions from mvgam models

Usage

plot_mvgam_uncertainty(
  object,
  series = 1,
  newdata,
  data_test,
  legend_position = "topleft",
  hide_xlabels = FALSE
)

Arguments

object: list object returned from mvgam. See mvgam()
series: integer specifying which series in the set is to be plotted
newdata: A dataframe or list containing at least 'series' and 'time' for the forecast horizon, in addition to any other variables included in the linear predictor of formula
data_test: Deprecated. Still works in place of newdata but users are recommended to use newdata instead for more seamless integration into R workflows
legend_position: The location may also be specified by setting x to a single keyword from the list: "none", "bottomright", "bottom", "bottomleft", "left", "topleft", "top", "topright", "right" and "center". This places the legend on the inside of the plot frame at the given location (if it is not "none").
hide_xlabels: logical. If TRUE, no xlabels are printed to allow the user to add custom labels using axis from base R

Value

A base R graphics plot

Details

The basic idea of this function is to compute forecasts by ignoring one of the two primary components in a correlated residual model (i.e. by either ignoring the linear predictor effects or by ignoring the residual dynamics). Some caution is required however, as this function was designed early in the mvgam development cycle and there are now many types of models that it cannot handle very well. For example, models with shared latent states, or any type of State-Space models that include terms in the trend_formula, will either fail or give nonsensical results. Improvements are in the works to provide a more general way to decompose forecast uncertainties, so please check back at a later date.