mcmcr is an R package to manipulate Monte Carlo Markov Chain (MCMC)
samples (Brooks et al. 2011).
To install the latest release from CRAN
To install the developmental version from GitHub
# install.packages("remotes") remotes::install_github("poissonconsulting/mcmcr")
For the purposes of this discussion, an MCMC sample represents the value of a term from a single iteration of a single chain. While a simple parameter such as an intercept corresponds to a single term, more complex parameters such as an interaction between two factors consists of multiple terms with their own inherent dimensionality - in this case a matrix. A set of MCMC samples can be stored in different ways.
The three most common S3 classes store MCMC samples as follows:
coda::mcmcstores the MCMC samples from a single chain as a matrix where each each row represents an iteration and each column represents a variable
mcmcobjects (with identical dimensions) as a list where each object represents a parallel chain
rjags::mcarraystores the samples from a single parameter where the initial dimensions are the parameter dimensions, the second to last dimension is iterations and the last dimension is chains.
In the first two cases the terms/parameters are represented by a single
dimension which means that the dimensionality inherent in the parameters
is stored in the labelling of the variables, ie,
"bInteraction[1,2]", "bInteraction[2,1]", .... The structure of the
mcmc.list objects emphasizes the time-series nature of MCMC
samples and is optimized for thining. In contrast
preserve the dimensionality of the parameters.
mcmcr package defines three related S3 classes which also preserve
the dimensionality of the parameters:
mcmcr::mcmcarrayis very similar to
rjags::mcarrayexcept that the first dimension is the chains, the second dimension is iterations and the subsequent dimensions represent the dimensionality of the parameter (it is called
mcmcarrayto emphasize that the MCMC dimensions ie the chains and iterations come first);
mcmcr::mcmcrstores multiple uniquely named
mcmcarrayobjects with the same number of chains and iterations.
mcmcrobjects with the same parameters, chains and iterations.
All five classes (
mcmcrs) are collectively referred to as MCMC objects.
mcmcarray objects were developed to facilitate manipulation of the
mcmcr objects were developed to allow a set of
dimensionality preserving parameters from a single analysis to be
manipulated as a whole.
mcmcrs objects were developed to allow the
results of multiple analyses using the same model to be manipulated
mcmcr package (together with the
nlist packages) introduces
a variety of (often) generic functions to manipulate and query
mcmcrs objects (and
In particular it provides functions to
coeftable (as a tibble);
nlist::nsamsas well as it’s parameter dimensions (
term::pdims) and term indices (
subsetobjects by chains, iterations and/or parameters;
bind_xxa pair of objects by their
combine_samples_n) or combine the samples of a single MCMC object by reducing its dimensions using
split_chainsan object’s chains;
mcmc_mapover an objects values;
esr(effectively sampling rate);
ess(effective sample size),
plotetc said objects.
The code is opinionated which has the advantage of providing a small set
of stream-lined functions. For example the only ‘convergence’ metric is
the uncorrected, untransformed, univariate split R-hat (potential scale
reduction factor). If you can convince me that additional features are
important I will add them or accept a pull request (see below).
Alternatively you might want to use the
mcmcr package to manipulate
your samples before coercing them to an
mcmc.list to take advantage of
all the summary functions in packages such as
library(mcmcr) mcmcr_example #> $alpha #>  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 #>  0.7911975 #> #> nchains: 2 #> niters: 400 coef(mcmcr_example, simplify = TRUE) #> term estimate lower upper svalue #> 1 alpha 3.7180250 2.2120540 5.232403 9.645658 #> 2 alpha 4.7180250 3.2120540 6.232403 9.645658 #> 3 beta[1,1] 0.9716535 0.2514796 1.713996 5.397731 #> 4 beta[2,1] 1.9716535 1.2514796 2.713996 7.323730 #> 5 beta[1,2] 1.9716535 1.2514796 2.713996 7.323730 #> 6 beta[2,2] 2.9716535 2.2514796 3.713996 9.645658 #> 7 sigma 0.7911975 0.4249618 2.559520 9.645658 rhat(mcmcr_example, by = "term") #> $alpha #>  2.002 2.002 #> #> $beta #> [,1] [,2] #> [1,] 1.147 1.147 #> [2,] 1.147 1.147 #> #> $sigma #>  1 plot(mcmcr_example[["alpha"]])
Please report any issues.
Pull requests are always welcome.
Please note that the mcmcr 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. ISBN: 978-1-4200-7941-8.
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.