R/risk.port2.R

In riskR: Risk Management

risk.port2 <-
function (x, alpha = c(0.05), beta = 1, p = 2) 
{
 m = dim(x)[2]
 P = array(rep(0, 11 * length(alpha) * dim(x)[2]), dim = c(11, length(alpha), dim(x)[2]))
  for (j in 1:11) {
   for (k in 1:length(alpha)) {
    f <- function(w = rep(1/m, m)) 
    {
     mean(as.vector(abs(w) %*% t(x)))/risk(as.vector(abs(w) %*% t(x)), alpha, beta, p)[j,k]
    }
    W = optim(rep(1/m, m), f,control = list(fnscale = -1))$par
    P[j, k, ] = abs(W)/sum(abs(W))
   }
  }
  colnames(P) <- paste(round(100 * alpha, 2), "%", sep = "")
  rownames(P) <- c("StD", "VaR", "EL", "ELD", "ES", "SDR", "EVaR", "DEVaR", "ENT", "DENT", "ML")
  return(P)
}