Revision 18b6abd7720cf34e0f34e4c63478dfca33bacea6 authored by Bendix Carstensen on 10 January 2007, 00:00:00 UTC, committed by Gabor Csardi on 10 January 2007, 00:00:00 UTC
1 parent b69dff2
ci.pd.Rd
\name{ci.pd}
\alias{ci.pd}
\title{
Compute confidence limits for a difference of two independent proportions.
}
\description{
The usual formula for the c.i. of at difference of proportions is
inaccurate. Newcombe has compared 11 methods and method 10 in his
paper looks like a winner. It is implemented here.
}
\usage{
ci.pd(aa, bb=NULL, cc=NULL, dd=NULL,
method = "Nc",
alpha = 0.05, conf.level=0.95,
digits = 3,
print = TRUE,
detail.labs = FALSE )
}
\arguments{
\item{aa}{Numeric vector of successes in sample 1. Can also be a
matrix or array (see details).}
\item{bb}{Successes in sample 2.}
\item{cc}{Failures in sample 1.}
\item{dd}{Failures in sample 2.}
\item{method}{Method to use for calculation of confidence interval, see
"Details".}
\item{alpha}{Significance level}
\item{conf.level}{Confidence level}
\item{print}{Should an account of the two by two table be printed.}
\item{digits}{How many digits should the result be rounded to if printed.}
\item{detail.labs}{Should the computing of probability differences be
reported in the labels.}
}
\details{
Implements method 10 from Newcombe(1998) (method="Nc") or from
Agresti & Caffo(2000) (method="AC").
\code{aa}, \code{bb}, \code{cc} and \code{dd} can be vectors.
If \code{aa} is a matrix, the elements \code{[1:2,1:2]} are used, with
successes \code{aa[,1:2]}. If \code{aa} is a three-way table or array,
the elements \code{aa[1:2,1:2,]} are used.
}
\value{
A matrix with three columns: probability difference, lower and upper
limit. The number of rows equals the length of the vectors \code{aa},
\code{bb}, \code{cc} and \code{dd} or, if \code{aa} is a 3-way matrix,
\code{dim(aa)[3]}.
}
\references{
RG Newcombe: Interval estimation for the difference between
independent proportions. Comparison of eleven methods. Statistics in
Medicine, 17, pp. 873-890, 1998.
A Agresti & B Caffo: Simple and effective confidence intervals for
proportions and differences of proportions result from adding two
successes and two failures. The American Statistician,
54(4), pp. 280-288, 2000.
}
\author{
Bendix Carstensen, Esa Läärä.
\url{http://www.biostat.ku.dk/~bxc}
}
\seealso{
\code{\link{twoby2}}, \code{\link{binom.test}}
}
\examples{
( a <- matrix( sample( 10:40, 4 ), 2, 2 ) )
ci.pd( a )
twoby2( t(a) )
prop.test( t(a) )
( A <- array( sample( 10:40, 20 ), dim=c(2,2,5) ) )
ci.pd( A )
ci.pd( A, detail.labs=TRUE, digits=3 )
}
\keyword{distribution}
\keyword{htest}
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...