R/snk.R

In ExpDes: Experimental Designs Package

#' Multiple comparison: Student-Newman-Keuls test
#'
#' \code{snk} Performs the test of SNK, for multiple
#' comparison of means.
#' @param y Numeric or complex vector containing the response
#' variable.
#' @param trt Numeric or complex vector containing the
#' treatments.
#' @param DFerror Error degrees of freedom.
#' @param SSerror Error sum of squares.
#' @param alpha Significance level.
#' @param group TRUE or FALSE.
#' @param main Title.
#' @return Returns the multiple comparison of means according
#' to the test of SNK.
#' @author Eric B Ferreira,
#'  \email{eric.ferreira@@unifal-mg.edu.br}
#' @author Denismar Alves Nogueira
#' @author Portya Piscitelli Cavalcanti
#' @seealso \code{\link{scottknott}}, \code{\link{duncan}},
#' \code{\link{lsd}}, \code{\link{lsdb}}, \code{\link{ccboot}},
#' \code{\link{tukey}}, \code{\link{ccF}}.
#' @examples
#' data(ex1)
#' attach(ex1)
#' crd(trat, ig, quali = TRUE, mcomp = "snk", sigT = 0.05)
#' @export

snk <-
function (y, trt, DFerror, SSerror, alpha = 0.05, group = TRUE,    main = NULL)
{
    MSerror <- SSerror/DFerror
    name.y <- paste(deparse(substitute(y)))
    name.t <- paste(deparse(substitute(trt)))
    junto <- subset(data.frame(y, trt), is.na(y) == FALSE)
    means <- tapply.stat(junto[, 1], junto[, 2], stat = "mean")
    sds <- tapply.stat(junto[, 1], junto[, 2], stat = "sd")
    nn <- tapply.stat(junto[, 1], junto[, 2], stat = "length")
    means <- data.frame(means, std.err = sds[, 2]/sqrt(nn[, 2]),
        replication = nn[, 2])
    names(means)[1:2] <- c(name.t, name.y)
    ntr <- nrow(means)

    # criando um contador de medias abrangidas para se calcular as prob de Tukey
    k.snk<-ntr-1
    Tprob <- vector(mode="integer",k.snk)
    kk <- 1
    for (kk in 1:k.snk)  {
        Tprob[kk] <- qtukey(1 - alpha, kk+1, DFerror)
        }
    # Tprob <- qtukey(1 - alpha, ntr, DFerror)

    nr <- unique(nn[, 2])
    nfila <- c("Alpha", "Error Degrees of Freedom", "Error Mean Square")
    nfila1 <- c("Distances between averages", "Critical Value of Studentized Range")
    # coloquei o vetor Tprop transposto
    nvalor <- c(alpha, DFerror, MSerror)
    nvalor1 <- rbind(t(seq(2,ntr)),t(Tprob))

    #cat("\nStudy:", main)
    #cat("\n\nHSD Test for", name.y, "\n")
    xtabla <- data.frame(...... = nvalor)
    xtabla1 <- data.frame(...... = nvalor1)
    row.names(xtabla) <- nfila
    row.names(xtabla1) <- nfila1
    #print(xtabla1)
    #print(xtabla)
    #cat("\nTreatment Means\n")
    #print(data.frame(row.names = NULL, means))
    HSD <- vector(mode="integer",k.snk)
    if (group) {
        if (length(nr) == 1) {
            kk <- 1
            for (kk in 1:k.snk) {
              HSD[kk] <- Tprob[kk] * sqrt(MSerror/nr)
                 }
            #HSD <- Tprob[1] * sqrt(MSerror/nr)
            #cat("\nHonestly Significant Difference", HSD)
        }
        else {
            nr1 <- 1/mean(1/nn[, 2])
            # criado o for para se ter um vetor de DMS
            kk <- 1
            for (kk in 1:k.snk) {
              HSD[kk] <- Tprob[kk] * sqrt(MSerror/nr1)
                 }
            #cat("\nHonestly Significant Difference", HSD)
            #cat("\nHarmonic Mean of Cell Sizes ", nr1)
            #cat("\n\nDifferent HSD for each comparison")
        }
        cat("\nStudent-Newman-Keuls's test (SNK)\n------------------------------------------------------------------------")
        cat("\nGroups  Treatments  Means\n")
        #output <- order.group(means[, 1], means[, 2], means[,4], MSerror, Tprob, means[, 3], parameter = 0.5)
        output <- order.stat.SNK(means[, 1], means[, 2], HSD)
        cat('------------------------------------------------------------------------\n')
    }
    #if (!group) {
    #    comb <- combn(ntmr, 2)
    #    nn <- ncol(comb)
    #    dif <- rep(0, nn)
    #    pvalue <- rep(0, nn)
    #    for (k in 1:nn) {
    #        i <- comb[1, k]
    #        j <- comb[2, k]
    #        dif[k] <- abs(means[i, 2] - means[j, 2])
    #        sdtdif <- sqrt(MSerror * (1/means[i, 4] + 1/means[j,
    #            4]))
    #        pvalue[k] <- round(1 - ptukey(dif[k] * sqrt(2)/sdtdif,
    #            ntr, DFerror), 4)
    #    }
    #    tr.i <- comb[1, ]
    #    tr.j <- comb[2, ]
    #    #cat("\nComparison between treatments means\n\n")
    #    #print(data.frame(row.names = NULL, tr.i, tr.j, diff = dif, pvalue = pvalue))
    #    output <- data.frame(trt = means[, 1], means = means[,2], M = "", N = means[, 4], std.err = means[, 3])
    #}
#    return(output)
}