[Returnanalytics-commits] r3828 - in pkg/Dowd: R man
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Fri Jul 17 16:35:12 CEST 2015
Author: dacharya
Date: 2015-07-17 16:35:12 +0200 (Fri, 17 Jul 2015)
New Revision: 3828
Added:
pkg/Dowd/R/LogNormalVaRETLPlot2DCL.R
pkg/Dowd/man/LogNormalVaRETLPlot2DCL.Rd
Log:
Function LogNormalVaRETLPlot2DCL added.
Added: pkg/Dowd/R/LogNormalVaRETLPlot2DCL.R
===================================================================
--- pkg/Dowd/R/LogNormalVaRETLPlot2DCL.R (rev 0)
+++ pkg/Dowd/R/LogNormalVaRETLPlot2DCL.R 2015-07-17 14:35:12 UTC (rev 3828)
@@ -0,0 +1,146 @@
+#' Plots log normal VaR and ETL against confidence level
+#'
+#' Plots the VaR and ETL of a portfolio against confidence level assuming that geometric
+#' returns are normally distributed, for specified confidence level and
+#' holding period.
+#'
+#' @param returns Vector of daily geometric return data
+#' @param mu Mean of daily geometric return data
+#' @param sigma Standard deviation of daily geometric return data
+#' @param investment Size of investment
+#' @param cl VaR confidence level and must be a vector
+#' @param hp VaR holding period and must be a scalar
+#'
+#' @note The input arguments contain either return data or else mean and
+#' standard deviation data. Accordingly, number of input arguments is either 4
+#' or 5. In case there are 4 input arguments, the mean and standard deviation of
+#' data is computed from return data. See examples for details.
+#'
+#' @references Dowd, K. Measuring Market Risk, Wiley, 2007.
+#'
+#' @author Dinesh Acharya
+#' @examples
+#'
+#' # Plots VaR and ETL against confidene level given geometric return data
+#' data <- runif(5, min = 0, max = .2)
+#' LogNormalVaRETLPlot2DCL(returns = data, investment = 5, cl = seq(.85,.99,.01), hp = 60)
+#'
+#' # Computes VaR against confidence level given mean and standard deviation of return data
+#' LogNormalVaRETLPlot2DCL(mu = .012, sigma = .03, investment = 5, cl = seq(.85,.99,.01), hp = 40)
+#'
+#'
+#' @export
+LogNormalVaRETLPlot2DCL<- function(...){
+ # Determine if there are four or five arguments, and ensure that arguments are read as intended
+ if (nargs() < 4) {
+ stop("Too few arguments")
+ }
+ if (nargs() > 5) {
+ stop("Too many arguments")
+ }
+ args <- list(...)
+ if (nargs() == 5) {
+ mu <- args$mu
+ investment <- args$investment
+ cl <- args$cl
+ sigma <- args$sigma
+ hp <- args$hp
+ }
+ if (nargs() == 4) {
+ mu <- mean(args$returns)
+ investment <- args$investment
+ cl <- args$cl
+ sigma <- sd(args$returns)
+ hp <- args$hp
+ }
+
+ # Check that inputs have correct dimensions
+ mu <- as.matrix(mu)
+ mu.row <- dim(mu)[1]
+ mu.col <- dim(mu)[2]
+ if (max(mu.row, mu.col) > 1) {
+ stop("Mean must be a scalar")
+ }
+ sigma <- as.matrix(sigma)
+ sigma.row <- dim(sigma)[1]
+ sigma.col <- dim(sigma)[2]
+ if (max(sigma.row, sigma.col) > 1) {
+ stop("Standard deviation must be a scalar")
+ }
+ cl <- as.matrix(cl)
+ cl.row <- dim(cl)[1]
+ cl.col <- dim(cl)[2]
+ if (min(cl.row, cl.col) > 1) {
+ stop("Confidence level must be a vector")
+ }
+ hp <- as.matrix(hp)
+ hp.row <- dim(hp)[1]
+ hp.col <- dim(hp)[2]
+ if (max(hp.row, hp.col) > 1) {
+ stop("Holding period must be a scalar")
+ }
+
+ # Check that cl is read as row vector
+ if (cl.row > cl.col) {
+ cl <- t(cl)
+ }
+
+ # Check that inputs obey sign and value restrictions
+ if (sigma < 0) {
+ stop("Standard deviation must be non-negative")
+ }
+ if (max(cl) >= 1){
+ stop("Confidence level(s) must be less than 1")
+ }
+ if (min(cl) <= 0){
+ stop("Confidence level(s) must be greater than 0")
+ }
+ if (min(hp) <= 0){
+ stop("Holding period must be greater than 0")
+ }
+ # VaR estimation
+ cl.row <- dim(cl)[1]
+ cl.col <- dim(cl)[2]
+ VaR <- investment - exp(sigma[1,1] * sqrt(hp[1,1]) * qnorm(1 - cl, 0, 1)+mu[1,1]*hp[1,1]*matrix(1,cl.row,cl.col) + log(investment)) # VaR
+
+ # ES estimation
+ n <- 1000 # Number of slices into which tail is divided
+ cl0 <- cl # Initial confidence level
+ delta.cl <- (1 - cl) / n # Increment to confidence level as each slice is taken
+ v <- VaR
+ for (i in 1:(n-1)) {
+ cl <- cl0 + i * delta.cl # Revised cl
+ v <- v + investment - exp(sigma[1,1] * sqrt(hp[1,1]) *
+ qnorm(1 - cl, 0, 1) + mu[1,1] * hp[1,1] *
+ matrix(1, cl.row, cl.col) + log(investment))
+ }
+ v <- v/n
+
+
+ # Plotting
+ ymin <- min(VaR, v)
+ ymax <- max(VaR, v)
+ xmin <- min(cl0)
+ xmax <- max(cl0)
+
+ plot(cl0, VaR, type = "l", xlim = c(xmin, xmax), ylim = c(ymin, ymax), xlab = "Confidence level", ylab = "VaR/ETL")
+ par(new=TRUE)
+ plot(cl0, v, type = "l", xlim = c(xmin, xmax), ylim = c(ymin, ymax), xlab = "Confidence level", ylab = "VaR/ETL")
+
+ title("Lognormal VaR and ETL against confidence level")
+ xmin <- min(cl0)+.3*(max(cl0)-min(cl0))
+ text(xmin,max(VaR)-.1*(max(VaR)-min(VaR)),
+ 'Input parameters', cex=.75, font = 2)
+ text(xmin,max(VaR)-.15*(max(VaR)-min(VaR)),
+ paste('Daily mean geometric return = ',round(mu[1,1],3)),cex=.75)
+ text(xmin,max(VaR)-.2*(max(VaR)-min(VaR)),
+ paste('Stdev. of daily geometric returns = ',round(sigma[1,1],3)),cex=.75)
+ text(xmin,max(VaR)-.25*(max(VaR)-min(VaR)),
+ paste('Investment size = ',investment),cex=.75)
+ text(xmin,max(VaR)-.3*(max(VaR)-min(VaR)),
+ paste('Holding period = ',hp,'days'),cex=.75)
+ # VaR and ETL labels
+ text(max(cl0)-.4*(max(cl0)-min(cl0)),min(VaR)+.3*(max(VaR)-min(VaR)),'Upper line - ETL',cex=.75);
+ text(max(cl0)-.4*(max(cl0)-min(cl0)),min(VaR)+.2*(max(VaR)-min(VaR)),'Lower line - VaR',cex=.75);
+
+}
Added: pkg/Dowd/man/LogNormalVaRETLPlot2DCL.Rd
===================================================================
--- pkg/Dowd/man/LogNormalVaRETLPlot2DCL.Rd (rev 0)
+++ pkg/Dowd/man/LogNormalVaRETLPlot2DCL.Rd 2015-07-17 14:35:12 UTC (rev 3828)
@@ -0,0 +1,47 @@
+% Generated by roxygen2 (4.1.1): do not edit by hand
+% Please edit documentation in R/LogNormalVaRETLPlot2DCL.R
+\name{LogNormalVaRETLPlot2DCL}
+\alias{LogNormalVaRETLPlot2DCL}
+\title{Plots log normal VaR and ETL against confidence level}
+\usage{
+LogNormalVaRETLPlot2DCL(...)
+}
+\arguments{
+\item{returns}{Vector of daily geometric return data}
+
+\item{mu}{Mean of daily geometric return data}
+
+\item{sigma}{Standard deviation of daily geometric return data}
+
+\item{investment}{Size of investment}
+
+\item{cl}{VaR confidence level and must be a vector}
+
+\item{hp}{VaR holding period and must be a scalar}
+}
+\description{
+Plots the VaR and ETL of a portfolio against confidence level assuming that geometric
+ returns are normally distributed, for specified confidence level and
+ holding period.
+}
+\note{
+The input arguments contain either return data or else mean and
+ standard deviation data. Accordingly, number of input arguments is either 4
+ or 5. In case there are 4 input arguments, the mean and standard deviation of
+ data is computed from return data. See examples for details.
+}
+\examples{
+# Plots VaR and ETL against confidene level given geometric return data
+ data <- runif(5, min = 0, max = .2)
+ LogNormalVaRETLPlot2DCL(returns = data, investment = 5, cl = seq(.85,.99,.01), hp = 60)
+
+ # Computes VaR against confidence level given mean and standard deviation of return data
+ LogNormalVaRETLPlot2DCL(mu = .012, sigma = .03, investment = 5, cl = seq(.85,.99,.01), hp = 40)
+}
+\author{
+Dinesh Acharya
+}
+\references{
+Dowd, K. Measuring Market Risk, Wiley, 2007.
+}
+
More information about the Returnanalytics-commits
mailing list