R/online_autoplot.R

In profoc: Probabilistic Forecast Combination Using CRPS Learning

#' Autoplot method for online models
#'
#' Plots the most recent weights in each quantile using ggplot2.
#' @param object Object of class inheriting from 'online'
#' @param ...  further arguments are ignored
#' @importFrom utils installed.packages
#' @export
autoplot.online <- function(object, ...) {
    if (requireNamespace("ggplot2", quietly = TRUE)) {
        weights <- object$weights[nrow(object$weights), , , , drop = FALSE]
        weights <- adrop(weights, 1)
        p <- object$specification$data$tau
        for (d in 1:dim(weights)[1]) {
            weight <- matrix(weights[d, , ])
            expert_names <- dimnames(object$experts_loss)[[4]]
            Expert <- as.character(rep(expert_names,
                each = ncol(weights)
            ))
            df <- data.frame(weight, Expert, p)
            fig <- ggplot2::ggplot(df, ggplot2::aes(x = p, y = weight, fill = Expert)) +
                ggplot2::theme_minimal() +
                ggplot2::geom_area() +
                ggplot2::ggtitle("Most Recent Combination Weights")
            return(fig)
        }
    } else {
        cat("Package ggplot2 needs to be installed to use autoplot.")
    }
}