yuen:

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

Usage

yuen(x, y = NULL, tr = 0.2, alpha = 0.05)

Arguments

x
y
tr
alpha

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 (x, y = NULL, tr = 0.2, alpha = 0.05) 
{
    if (is.null(y)) {
        if (is.matrix(x) || is.data.frame(x)) {
            y = x[, 2]
            x = x[, 1]
        }
        if (is.list(x)) {
            y = x[[2]]
            x = x[[1]]
        }
    }
    if (tr == 0.5) 
        stop("Using tr=.5 is not allowed; use a method designed for medians")
    if (tr > 0.25) 
        print("Warning: with tr>.25 type I error control might be poor")
    x <- x[!is.na(x)]
    y <- y[!is.na(y)]
    h1 <- length(x) - 2 * floor(tr * length(x))
    h2 <- length(y) - 2 * floor(tr * length(y))
    q1 <- (length(x) - 1) * winvar(x, tr)/(h1 * (h1 - 1))
    q2 <- (length(y) - 1) * winvar(y, tr)/(h2 * (h2 - 1))
    df <- (q1 + q2)^2/((q1^2/(h1 - 1)) + (q2^2/(h2 - 1)))
    crit <- qt(1 - alpha/2, df)
    dif <- mean(x, tr) - mean(y, tr)
    low <- dif - crit * sqrt(q1 + q2)
    up <- dif + crit * sqrt(q1 + q2)
    test <- abs(dif/sqrt(q1 + q2))
    yuen <- 2 * (1 - pt(test, df))
    list(n1 = length(x), n2 = length(y), est.1 = mean(x, tr), 
        est.2 = mean(y, tr), ci = c(low, up), p.value = yuen, 
        dif = dif, se = sqrt(q1 + q2), teststat = test, crit = crit, 
        df = df)
  }

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