mcmcderive
is an R package to generate derived parameter(s) from Monte
Carlo Markov Chain (MCMC) samples using R code.
This is useful because it means Bayesian models can be fitted without the inclusion of derived parameters which add unnecessary clutter and slows model fitting. For more information on MCMC samples see Brooks et al. (2011).
library(mcmcderive)
#> Registered S3 method overwritten by 'mcmcr':
#> method from
#> as.mcmc.nlists nlist
mcmcr::mcmcr_example
#> $alpha
#> [1] 3.718025 4.718025
#>
#> nchains: 2
#> niters: 400
#>
#> $beta
#> [,1] [,2]
#> [1,] 0.9716535 1.971654
#> [2,] 1.9716535 2.971654
#>
#> nchains: 2
#> niters: 400
#>
#> $sigma
#> [1] 0.7911975
#>
#> nchains: 2
#> niters: 400
expr <- "
log(alpha2) <- alpha
gamma <- sum(alpha) * sigma
"
mcmc_derive(mcmcr::mcmcr_example, expr, silent = TRUE)
#> $alpha2
#> [1] 41.18352 111.94841
#>
#> nchains: 2
#> niters: 400
#>
#> $gamma
#> [1] 6.60742
#>
#> nchains: 2
#> niters: 400
If the MCMC object has multiple chains the run time can be substantially reduced by generating the derived parameters for each chain in parallel. In order for this to work it is necessary to:
- Ensure plyr and doParallel are installed using
install.packages(c("plyr", "doParallel"))
. - Register a parallel backend using
doParallel::registerDoParallel(4)
. - Set
parallel = TRUE
in the call tomcmc_derive()
.
To facilitate the translation of model code into R code the extras
package provides the R equivalent to common model functions such as
pow()
, phi()
and log() <-
.
To install the release version from CRAN.
install.packages("mcmcderive")
The website for the release version is at https://poissonconsulting.github.io/mcmcderive/.
To install the development version from GitHub
# install.packages("remotes")
remotes::install_github("poissonconsulting/mcmcderive")
or from r-universe.
install.packages("mcmcderive", repos = c("https://poissonconsulting.r-universe.dev", "https://cloud.r-project.org"))
Please report any issues.
Pull requests are always welcome.
Please note that the mcmcderive project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.
Brooks, S., Gelman, A., Jones, G.L., and Meng, X.-L. (Editors). 2011. Handbook for Markov Chain Monte Carlo. Taylor & Francis, Boca Raton.