CompClassicDist:

In musto101/wilcox_R: What the package does (short line)

Usage

CompClassicDist(svd, out)

Arguments

svd
out

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 (svd, out) 
{
    out$classic$P <- as.matrix(svd$loadings[, 1:out$k])
    out$classic$L <- as.vector(svd$eigenvalues[1:out$k])
    out$classic$M <- as.vector(svd$centerofX)
    out$classic$T <- as.matrix(svd$scores[, 1:out$k])
    out$classic$k <- out$k
    out$classic$Xc <- as.matrix(svd$Xcentered)
    Tclas <- out$classic$Xc %*% out$classic$P
    out$classic$sd <- sqrt(mahalanobis(Tclas, center = rep(0, 
        length = ncol(Tclas)), cov = diag(x = out$classic$L, 
        nrow = out$classic$k)))
    out$classic$cutoff$sd <- sqrt(qchisq(0.975, out$classic$k))
    Xtilde <- Tclas %*% t(out$classic$P)
    Cdiff <- out$classic$Xc - Xtilde
    out$classic$od <- vector(mode = "numeric", length = nrow(out$classic$Xc))
    if (is.list(dimnames(out$classic$T))) {
        names(out$classic$od) <- dimnames(out$classic$T)[[1]]
    }
    out$classic$od <- apply(Cdiff, 1, vecnorm)
    if (out$k != svd$rank) {
        m <- mean(out$classic$od^(2/3))
        s <- sqrt(var(out$classic$od^(2/3)))
        out$classic$cutoff$od <- sqrt((qnorm(0.975, m, s))^3)
    }
    else {
        out$classic$cutoff$od <- 0
    }
    out$classic$flag <- (out$classic$od <= rep(x = out$classic$cutoff$od, 
        times = length(out$classic$od))) & (out$classic$sd <= 
        rep(x = out$classic$cutoff$sd, times = length(out$classic$sd)))
    out$classic$class <- "CPCA"
    return(out)
  }

musto101/wilcox_R documentation built on May 23, 2019, 10:52 a.m.