Functions that compute probabilistic excursion sets, contour credibility regions, contour avoiding regions, and simultaneous confidence bands for latent Gaussian random processes and fields. The package also contains functions that calculate these quantities for models estimated with the INLA package. The main references for excursions are Bolin and Lindgren (2015)
Excursions is an R package that contains functions that compute probabilistic excursion sets, contour credibility regions, and simultaneous confidence bands for latent Gaussian random processes and fields.
The theory of the methods used in the package are described in Bolin and Lindgren (2014). A manual for the package can be found here and the code used in the manual can be downloaded here.
The current stable version of the package is 1.1, which is also on CRAN. The development version of the package contains new features and fixes that are not on CRAN. To install this version, see the instructions below.
The stable version of the package can be installed directly from CRAN, or by using the command
#!r devtools::install_bitbucket("excursions","davidbolin",ref="default")
in R. The development version can be installed using the command
#!r devtools::install_bitbucket("excursions","davidbolin",ref="devel")
The development version of the package is called excursionsdevel, so it is loaded by running
#!rlibrary(excursionsdevel)
If you want to install the package using the install_bitbucket-method on Windows, you first need to install Rtools and add the paths to Rtools and gcc to the Windows PATH environment variable. This can be done for the current R session only using the commands
#!rrtools = "C:\\Rtools\\bin"gcc = "C:\\Rtools\\gcc-4.6.3\\bin"Sys.setenv(PATH=paste(c(gcc,rtools, Sys.getenv("PATH")),collapse=";"))
where the variables rtools and gcc need to be changed if Rtools is not installed directly on C:.
Added a NEWS.md file to track changes to the package.
Updated repository for INLA
Add n.iter and seed options for contourmap() calculations.
Add support for the QC method for contourmap.inla().
Updated documentation.
gaussint().excursions.variances() caused by the previous fix to gaussint().gaussint() that affected standalone use only.rBind and cBind and add dependency on R >= 3.2.0Add whitespace after -f for make, for wider compatibility, see
http://pubs/opegroup.org/onlinepubs/9699919799/utilities/make.html
Updated CITATION information