[R] Normal probability plot of effects

[kjetil halvorsen]

I have a local function:

> qqnorm.aov
function (obj, full = FALSE, label = FALSE, omit = NULL, xlab = paste(if
(full) "" else "Half", 
    " Normal plot"), ylab = "Effects", ...) 
{
    r <- obj$rank
    eff <- if (full) 
        effects(obj, set.sign = TRUE)[1:r]
    else abs(effects(obj))[1:r]
    na <- names(eff)
    int <- match("(Intercept)", na)
    if (!is.null(omit)) {
        if (is.character(omit)) {
            int <- c(int, match(omit, na))
        }
        else int <- c(int, omit)
    }
    int <- int[!is.na(int)]
    if (length(int)) 
        eff <- eff[-int]
    n <- length(eff)
    if (n <= 0) 
        stop("Not enough effects")
    ord <- order(eff)
    na <- names(eff)
    P <- if (full) 
        ppoints(n)
    else ((1:n) + n)/(2 * n + 1)
    Q <- qnorm(P)
    plot(x = Q, y = eff[ord], xlab = xlab, ylab = ylab, ...)
    if (label && dev.interactive()) 
        identify(Q, eff[ord], names(eff)[ord])
}


which is an incomplete implementation of the s function with the same
name. It could help. (It is of course most usefull for orthogonal
designs)

Kjetil Halvorsen


Kyriakos Kachrimanis wrote:
> 
> Can anybody help me on this, please:
> 
> I want to produce a normal probability plot of effects for an unreplicated
> design with 4 factors. 2 of the factors have 2 levels and the other 2 have 4
> levels. Is it possible to produce such a graph?
> I want to use this graph to decide which interaction effects are negligible
> to combine them as an estimate of error.
> Thank you in advance.
> 
> Kyriakos Kachrimanis
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._