https://github.com/cran/unmarked
Tip revision: 0e9915b1bbee346e4c283f39772af69032684e39 authored by Ken Kellner on 09 January 2024, 10:20:02 UTC
version 1.4.1
version 1.4.1
Tip revision: 0e9915b
unmarkedFrame.Rd
\name{unmarkedFrame}
\title{Create an unmarkedFrame, or one of its child classes.}
\alias{unmarkedFrame}
\usage{unmarkedFrame(y, siteCovs=NULL, obsCovs=NULL, mapInfo, obsToY)}
\description{Constructor for unmarkedFrames.}
\arguments{
\item{y}{An MxJ matrix of the observed measured data, where M is the
number of sites and J is the maximum number of observations per site.}
\item{siteCovs}{A \code{\link{data.frame}} of covariates that vary at the
site level. This should have M rows and one column per covariate}
\item{obsCovs}{Either a named list of \code{\link{data.frame}}s of
covariates that vary within sites, or a \code{\link{data.frame}}
with MxJ rows in site-major order.}
\item{obsToY}{optional matrix specifying relationship between
observation-level covariates and response matrix}
\item{mapInfo}{geographic coordinate information. Currently ignored.}
}
\details{
unmarkedFrame is the S4 class that holds data structures to be passed
to the model-fitting functions in unmarked.
An unmarkedFrame contains the observations (\code{y}), covariates
measured at the observation level (\code{obsCovs}), and covariates
measured at the site level (\code{siteCovs}).
For a data set with M sites and J observations at each site, y is an
M x J matrix. \code{obsCovs} and \code{siteCovs} are both data frames
(see \link{data.frame}). \code{siteCovs} has M rows so that each row
contains the covariates for the corresponding sites.
\code{obsCovs} has M*obsNum rows so that each covariates is ordered by
site first, then observation number. Missing values are coded with
\code{NA} in any of y, siteCovs, or obsCovs.
Additionally, unmarkedFrames contain metadata: obsToY, mapInfo.
obsToY is a matrix describing relationship between response matrix and
observation-level covariates. Generally this does not need to be
supplied by the user; however, it may be needed when using
\code{\link{multinomPois}}. For example, double observer sampling, y
has 3 columns corresponding the observer 1, observer 2, and both, but
there were only two independent observations.
In this situation, y has 3 columns, but obsToY must be specified.
Several child classes of \code{unmarkedFrame} require addional
metadata. For example, \code{unmarkedFrameDS} is used to organize
distsance sampling data for the \code{\link{distsamp}} function, and
it has arguments dist.breaks, tlength, survey, and unitsIn, which
specify the distance interval cut points, transect lengths, "line" or
"point" transect, and units of measure, respectively.
All site-level covariates are automatically copied to obsCovs so that
site level covariates are available at the observation level.
}
\value{an unmarkedFrame object}
\seealso{\code{\link{unmarkedFrame-class}},
\code{\link{unmarkedFrameOccu}}, \code{\link{unmarkedFramePCount}},
\code{\link{unmarkedFrameDS}}}
\examples{
# Set up data for pcount()
data(mallard)
mallardUMF <- unmarkedFramePCount(mallard.y, siteCovs = mallard.site,
obsCovs = mallard.obs)
summary(mallardUMF)
# Set up data for occu()
data(frogs)
pferUMF <- unmarkedFrameOccu(pfer.bin)
# Set up data for distsamp()
data(linetran)
ltUMF <- with(linetran, {
unmarkedFrameDS(y = cbind(dc1, dc2, dc3, dc4),
siteCovs = data.frame(Length, area, habitat),
dist.breaks = c(0, 5, 10, 15, 20),
tlength = linetran$Length * 1000, survey = "line", unitsIn = "m")
})
summary(ltUMF)
# Set up data for multinomPois()
data(ovendata)
ovenFrame <- unmarkedFrameMPois(ovendata.list$data,
siteCovs=as.data.frame(scale(ovendata.list$covariates[,-1])),
type = "removal")
summary(ovenFrame)
\dontrun{
# Set up data for colext()
frogUMF <- formatMult(masspcru)
summary(frogUMF)
}
}