R/empirical_stats.R

In Reckziegel/CMA: Copula Marginal Algorithm (CMA)

#' Summary Statistics for Empirical Distributions
#'
#' Computes the mean, standard deviation, skewness, kurtosis, Value-at-Risk (VaR)
#' and Conditional Value-at-Risk CVaR) under flexible probabilities.
#'
#' @param x A (non-tidy) scenario-probability distribution.
#' @param p A probability vector. If \code{NULL} (the default) the standard
#' 1 over N is used.
#' @param level A number with the desired probability level. The default is
#' \code{level = 0.01}.
#'
#' @return A \code{tibble} with 2 columns and 6 rows.
#' @export
#'
#' @examples
#' library(ggplot2)
#'
#' x <- diff(log(EuStockMarkets))
#'
#' calm_market  <- panic_copula(x = x, n = 1000, panic_prob = 0.00, dist = "normal")
#' panic_market <- panic_copula(x = x, n = 1000, panic_prob = 0.20, dist = "normal")
#'
#' # Plot Calm market stats
#' emp_calm <- empirical_stats(calm_market$simulation, p = calm_market$p)
#' ggplot(emp_calm, aes(x = name, y = value)) +
#'  geom_col() +
#'  facet_wrap(~stat, scales = "free_y") +
#'  scale_y_continuous(labels = scales::number_format(accuracy = 0.01))
#'
#' # Plot Panic market stats
#' emp_panic <- empirical_stats(panic_market$simulation, p = panic_market$p)
#' ggplot(emp_panic, aes(x = name, y = value)) +
#'  geom_col() +
#'  facet_wrap(~stat, scales = "free_y") +
#'  scale_y_continuous(labels = scales::number_format(accuracy = 0.01))
empirical_stats <- function(x, p = NULL, level = 0.01) {

    assertthat::assert_that(assertthat::is.number(level))

    ew <- rep(1 / NROW(x), NROW(x))
    ffp::empirical_stats(
      x     = x,
      p     = if (!is.null(p)) ffp::as_ffp(p) else ffp::as_ffp(ew),
      level = level
    )

}