dagostino.pearson.test <- function(x) {

    # from Zar (1999), implemented by Doug Scofield, scofield@bio.indiana.edu

    DNAME <- deparse(substitute(x))
    n <- length(x);

    x2 <- x * x;
    x3 <- x * x2;
    x4 <- x * x3;

    # compute Z_g1
    k3 <- ((n*sum(x3)) - (3*sum(x)*sum(x2)) + (2*(sum(x)^3)/n)) / ((n-1)*(n-2));

    g1 <- k3 / sqrt(var(x)^3);
    sqrtb1 <- ((n - 2)*g1) / sqrt(n*(n - 1));

    A <- sqrtb1 * sqrt(((n + 1)*(n + 3)) / (6*(n - 2)));
    B <- (3*(n*n + 27*n - 70)*(n+1)*(n+3)) / ((n-2)*(n+5)*(n+7)*(n+9));
    C <- sqrt(2*(B - 1)) - 1;
    D <- sqrt(C);
    E <- 1 / sqrt(log(D));
    F <- A / sqrt(2/(C - 1));

    Zg1 <- E * log(F + sqrt(F*F + 1));

    # compute Z_g2
    G <- (24*n*(n-2)*(n-3)) / (((n+1)^2)*(n+3)*(n+5));
    k4 <- (((n*n*n + n*n)*sum(x4)) - (4*(n*n + n)*sum(x3)*sum(x)) - (3*(n*n - n)*sum(x2)^2) + (12*n*sum(x2)*sum(x)^2) - (6*sum(x)^4)) / (n*(n-1)*(n-2)*(n-3));
    g2 <- k4 / var(x)^2;
    H <- ((n-2)*(n-3)*abs(g2)) / ((n+1)*(n-1)*sqrt(G));

    J <- ((6*(n*n - 5*n + 2)) / ((n+7)*(n+9))) * sqrt((6*(n+3)*(n+5)) / (n*(n-2)*(n-3)));
    K <- 6 + (8/J)*(2/J + sqrt(1 + 4/(J*J)));
    L <- (1 - 2/K) / (1 + H*sqrt(2/(K-4)));
    Zg2 <- (1 - 2/(9*K) - (L^(1/3))) / (sqrt(2/(9*K)));
    K2 <- Zg1*Zg1 + Zg2*Zg2;
    pk2 <- pchisq(K2, 2, lower.tail=FALSE);
    RVAL <- list(statistic = c(K2 = K2, g1=g1, g2=g2, Zg1=Zg1, Zg2=Zg2), p.value = pk2, method = "D'Agostino-Pearson normality test\n\nK2 is distributed as Chi-squared with df=2", alternative = "distribution is not normal", data.name = DNAME);
    class(RVAL) <- "htest";
    return(RVAL);
}
