Bayesian inference for state-space models with R

RBi is an R interface to libbi, a library for Bayesian inference.

It mainly contains:

• various functions to retrieve and process the results from libbi (which are in NetCDF format)
• a bi_model class, to manipulate libbi models
• a libbi wrapper class, to perform Bayesian using libbi inference from within R,

Installation

RBi requires R (>= 2.12.1) as well as the packages:

• reshape2
• ncdf4
• data.table

The easiest way to install the latest stable version of RBi is via CRAN. The package is called rbi (all lower case):

install.packages('rbi')

Alternatively, the current development version can be installed using the devtools package

# install.packages("devtools")
library('devtools')
install_github("libbi/rbi")

The RBi package has only been tested on GNU/Linux and OS X, but it should mostly work everywhere R works.

If you want to use RBi as a wrapper to LibBi then you need a working version of LibBi. To install LibBi on a Mac or Unix, the easiest way is to install Homebrew (on OS X) or Linuxbrew (on linux), followed by (using a command shell, i.e. Terminal or similar):

brew install libbi

The path to libbi script can be passed as an argument to RBi, otherwise the package tries to find it automatically using the which linux/unix command.

If you just want to process the output from LibBi, then you do not need to have LibBi installed.

Getting started

A good starting point is to look at the included demos:

 demo(PZ_generate_dataset) ## generating a data set from a model
 demo(PZ_PMMH)             ## particle Markov-chain Metropolis-Hastings
 demo(PZ_SMC2)             ## SMC^2
 demo(PZ_filtering)        ## filtering

For further information, have a look at the introductory vignette from the link from the rbi CRAN package.

Using coda

LibBi contains the get_traces method which provides an interface to coda.

Other packages

For higher-level methods to interact with LibBi, have a look at RBi.helpers.