Loading an R Package from a Custom directory

Try using Hadley Wickham's devtools package, which allows loading packages from a given directory:

library(devtools)# load package w/o installingload_all('/some/package/diR')# or invoke 'R CMD INSTALL'install('/some/package/diR')

Please add some extra information on the operating system. If you're on windows, you need Rtools ( http://www.murdoch-sutherland.com/Rtools/ ) to build from source. See that website for more information on how to install everything you need.

Even when you're on Linux, simply extracting the package-file doesn't work. There might be underlying C-code (which is the case for the MSBVAR package), and even R code has to be processed in order to be built into a package that can be loaded directly with the library() function.

Plus, you have to take into account that the package you want to install might have dependencies. For the MSBVAR package, these are the packages coda and bit. When building from source, you need to make sure all dependencies are installed as well, or you can get errors.

apart from the R CMD INSTALL you could try from within R :

# from CRANinstall.packages("MSBVAR", type="source")# from a local file install.packages("/my/dir/MSBVAR.tar.gz",repos=NULL, type="source")

or why not just do

# from CRANinstall.packages("MSBVAR")

This works perfectly fine.

You need to install the package to a directory to which you have permission to read and write. First, download the package to an easily accessible directory. If you're on Linux/Mac, try creating a directory called 'rlib' in your home directory.

cd ~; mkdir rlibR CMD INSTALL MSBVAR.tar.gz --library=rlib

If you would prefer to install the package from R, do this:

## From CRANinstall.packages("MSBVAR", lib="~/rlib")