R help - Jan 2009 - question about the scale in ridge regression of the MASS package

Hi

I am reading the source code of the ridge regression of the MASS package,
and I found the following piece of code
    X <- model.matrix(Terms, m, contrasts)
    n <- nrow(X); p <- ncol(X)
    offset <- model.offset(m)
    if(!is.null(offset)) Y <- Y - offset
    if(Inter <- attr(Terms, "intercept"))
    {
        Xm <- colMeans(X[, -Inter])
        Ym <- mean(Y)
        p <- p - 1
        X <- X[, -Inter] - rep(Xm, rep(n, p))
        Y <- Y - Ym
    } else Ym <- Xm <- NA
   * Xscale <- drop(rep(1/n, n) %*% X^2)^0.5*    # line 38 of the original
code
    X <- X/rep(Xscale, rep(n, p))

It uses sqrt((x-xbar)^2)/n to calculate the scale of the data, while the
scale function provided by R uses sqrt((x-xbar)^2)/(n-1), which calculates
the standard deviation of the data. Is this a mistake in the MASS package,
or there are some other reasons?

Thank you very much.

L.J.

	[[alternative HTML version deleted]]