ChainLadder (version 0.1.1-5)

Mack.S.E: ~~function to do ... ~~

Description

~~ A concise (1-5 lines) description of what the function does. ~~

Usage

Mack.S.E(MackModel, FullTriangle)

Arguments

MackModel

~~Describe MackModel here~~

FullTriangle

~~Describe FullTriangle here~~

Value

• ~Describe the value returned If it is a LIST, use
• comp1
  Description of 'comp1'
• comp2
  Description of 'comp2'
• ...

Details

~~ If necessary, more details than the description above ~~

References

~put references to the literature/web site here ~

See Also

~~objects to See Also as help, ~~~

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (MackModel, FullTriangle) 
{
    n <- ncol(FullTriangle)
    m <- nrow(FullTriangle)
    f <- rep(1, n)
    f.se <- rep(0, n)
    sigma <- rep(0, (n - 1))
    f[1:(n - 1)] <- sapply(MackModel, function(x) summary(x)$coef["x", 
        "Estimate"])
    f.se[1:(n - 1)] <- sapply(MackModel, function(x) summary(x)$coef["x", 
        "Std. Error"])
    sigma[1:(n - 1)] <- sapply(MackModel, function(x) summary(x)$sigma)
    dev = c(1:(n - 2))
    sigmaModel <- lm(log(sigma[-(n - 1)]) ~ dev)
    sigma[n - 1] <- exp(predict(sigmaModel, newdata = data.frame(dev = (n - 
        1))))
    f.se[n - 1] = sigma[n - 1]/sqrt(FullTriangle[1, n - 1])
    F.se <- t(t(1/sqrt(FullTriangle)[, -n]) * (sigma))
    FullTriangle.se <- FullTriangle * 0
    for (i in 2:m) {
        for (k in c((n + 1 - i):(n - 1))) {
            FullTriangle.se[i, k + 1] = sqrt(FullTriangle[i, 
                k]^2 * (F.se[i, k]^2 + f.se[k]^2) + FullTriangle.se[i, 
                k]^2 * f[k]^2)
        }
    }
    return(list(sigma = sigma, f = f, f.se = f.se, FullTriangle.se = FullTriangle.se))
  }