Dowd (version 0.12)

PCAVaR: Estimates VaR by principal components analysis

Description

Estimates the VaR of a multi position portfolio by principal components analysis, using chosen number of principal components and a specified confidence level or range of confidence levels.

Usage

PCAVaR(Ra, position.data, number.of.principal.components, cl)

Arguments

Ra

Matrix return data set where each row is interpreted as a set of daily observations, and each column as the returns to each position in a portfolio

position.data

Position-size vector, giving amount invested in each position

number.of.principal.components

Chosen number of principal components

cl

Chosen confidence level

Value

VaR

References

Dowd, K. Measuring Market Risk, Wiley, 2007.

Examples

# Computes PCA VaR
   Ra <- matrix(rnorm(4*6),4,6)
   position.data <- rnorm(6)
   PCAVaR(Ra, position.data, 2, .95)