https://github.com/cran/ape
Tip revision: 5d37575f69ed899da1eaba005d0267b54381ac23 authored by Emmanuel Paradis on 17 February 2012, 00:00:00 UTC
version 3.0-1
version 3.0-1
Tip revision: 5d37575
correlogram.formula.Rd
\name{correlogram.formula}
\alias{correlogram.formula}
\title{Phylogenetic Correlogram}
\usage{
correlogram.formula(formula, data = NULL, use = "all.obs")
}
\arguments{
\item{formula}{a formula of the type \code{y1+..+yn ~ g1/../gn}, where
the \code{y}'s are the data to analyse and the \code{g}'s are the
taxonomic levels.}
\item{data}{a data frame containing the variables specified in the
formula. If \code{NULL}, the variables are sought in the user's
workspace.}
\item{use}{a character string specifying how to handle missing
values (i.e., \code{NA}). This must be one of "all.obs",
"complete.obs", or "pairwise.complete.obs", or any unambiguous
abbrevation of these. In the first case, the presence of missing
values produces an error. In the second case, all rows with missing
values will be removed before computation. In the last case, missing
values are removed on a case-by-case basis.}
}
\description{
This function computes a correlogram from taxonomic levels.
}
\details{
See the vignette in R: \code{vignette("MoranI")}.
}
\value{
An object of class \code{correlogram} which is a data frame with three
columns:
\item{obs}{the computed Moran's I}
\item{p.values}{the corresponding P-values}
\item{labels}{the names of each level}
or an object of class \code{correlogramList} containing a list of
objects of class \code{correlogram} if several variables are given as
response in \code{formula}.
}
\author{Julien Dutheil \email{julien.dutheil@univ-montp2.fr} and
Emmanuel Paradis}
\seealso{
\code{\link{plot.correlogram}, \link{Moran.I}}
}
\examples{
data(carnivora)
### Using the formula interface:
co <- correlogram.formula(SW ~ Order/SuperFamily/Family/Genus,
data=carnivora)
co
plot(co)
### Several correlograms on the same plot:
cos <- correlogram.formula(SW + FW ~ Order/SuperFamily/Family/Genus,
data=carnivora)
cos
plot(cos)
}
\keyword{regression}