R/ci.sc.R

In mcfanda/cpower: Power analysis for contrasts

# The code used here is not originally developed for this package but it comes from:
# Package: MBESS
# Type: Package
# Title: The MBESS R Package
# Version: 4.4.2
# Date: 2017-12-19
# Authors@R: c(person("Ken", "Kelley", role=c("aut", "cre"), email="kkelley@nd.edu"))
# Maintainer: Ken Kelley <kkelley@nd.edu>
# the code has been slightly adjusted for the purposes of the present package


.ci.sc <- function(means=NULL, s.anova=NULL, c.weights=NULL, n=NULL, N=NULL,
Psi=NULL, ncp=NULL, conf.level=.95, alpha.lower=NULL, alpha.upper=NULL, df.error=NULL, ...)
{

if(!identical(sum(c.weights[c.weights>0]), 1)) stop("Please use fractions to specify the contrast weights")
if(!identical(round(sum(c.weights), 5), 0)) stop("The sum of the contrast weights ('c.weights') should equal zero.")

if(length(n)==1)
{
n <- rep(n, length(c.weights))
}

if(length(n)!=length(c.weights)) stop("The lengths of 'n' and 'c.weights' differ, which should not be the case.")


part.of.se <- sqrt(sum((c.weights^2)/n))

if(!is.null(Psi))
{
if(!is.null(means)) stop("Since the contrast effect ('Psi') was specified, you should not specify the vector of means ('means').")
if(!is.null(ncp)) stop("Since the contrast effect ('Psi') was specified, you should not specify the noncentral parameter ('ncp').")
if(is.null(s.anova)) stop("You must specify the standard deviation of the errors (i.e., the square root of the error variance).")
if(is.null(n)) stop("You must specify the vector per group/level sample size ('n').")
if(is.null(c.weights)) stop("You must specify the vector of contrast weights ('c.weights').")

psi <- Psi/s.anova
lambda <- psi*part.of.se
}

if(!is.null(ncp))
{
if(!is.null(means)) stop("Since the noncentral parameter was specified directly, you should not specify the vector of means ('means').")
if(!is.null(Psi)) stop("Since the noncentral parameter was specified directly, you should not specify the the contrast effect ('Psi').")
if(is.null(s.anova)) stop("You must specify the standard deviation of the errors (i.e., the square root of the error variance).")
if(is.null(n)) stop("You must specify the vector per group/level sample size ('n').")
if(is.null(c.weights)) stop("You must specify the vector of contrast weights ('c.weights'.")

lambda <- ncp
}

if(!is.null(means))
{
Psi <- sum(c.weights*means)
psi <- Psi/s.anova
lambda <- psi/part.of.se

}

if(is.null(alpha.lower) & is.null(alpha.upper))
{
alpha.lower <- (1-conf.level)/2
alpha.upper <- (1-conf.level)/2
}

if(is.null(N)) N=n*length(c.weights)
if(is.null(df.error)) df.2 <- N - length(c.weights)

Lims <- .conf.limits.nct(ncp=lambda, df=df.2, conf.level = NULL, alpha.lower = alpha.lower,
        alpha.upper = alpha.upper, sup.int.warns=TRUE, method = "all", ...)

Result <- list(Lower.Conf.Limit.Standardized.Contrast = Lims$Lower.Limit*part.of.se, Standardized.contrast = psi,
        Upper.Conf.Limit.Standardized.Contrast = Lims$Upper.Limit*part.of.se)

# print(paste("The", 1 - (alpha.lower + alpha.upper), "confidence limits for the standardized contrast are given as:"))

return(Result)
}