R/adup.test.R

In truncgof: GoF tests allowing for left truncated data

"adup.test" <-
function (x, distn, fit, H = NA, 
    alternative = c("two.sided", "less", "greater"), sim = 100, tol = 1e-04, estfun = NA) 
{
    if (!is.function(try(get(distn), silent = TRUE)))
       stop("'distn' must be a character string of a distribution function")
    if (is.na(H)) H <- -Inf
    est <- try(eval(parse(text = estfun)), silent = TRUE)    
    if (!is.na(est[1])) {
        if (class(est) == "try-error")
            stop(paste("can not evaluat 'estfun':\n", est[1]))
        if (!is.vector(est) || !is.list(est))
            stop("value of 'estfun' is not a vector of 'mode(list)'")
        if (!all(names(est) %in% names(formals(distn))))
            stop("values of 'estfun' dosn't match arguments in 'distn'")
    }
    adup.p <- function(n, zH, z, j) sqrt(n)*max((zH + j/n*(1-zH) - z)/(1-z))
    adup.m <- function(n, zH, z, j) sqrt(n)*max((z - zH - (j-1)/n * (1-zH))/(1-z))
    adup   <- function(n, zH, z, j) sqrt(n) * max(max((zH + j/n * (1-zH) - z)/(1-z)), 
              max((z - zH - (j-1)/n * (1-zH))/(1-z)))    
    AL <- match.arg(alternative)
    TS <- switch(AL, "two.sided" = adup, "less" = adup.m, "greater" = adup.p)
    val <- mctest(x, distn, fit, H, sim, tol, STATISTIC = TS, estfun)
    STATISTIC <- val$TS
    names(STATISTIC) <- switch(AL, two.sided = "ADup", less = "ADup-", greater = "ADup+")    
    res <- list(method = "Anderson-Darling Upper Tail Test", statistic = STATISTIC, 
        p.value = val$p.value, alternative = AL, data.name = deparse(substitute(x)), 
        sim.no = val$sim, treshold = H)
    class(res) <- "mchtest"
    return(res)
}