Bayesian Structure Learning in Graphical Models using Birth-Death MCMC

Provides statistical tools for Bayesian structure learning in undirected graphical models for continuous, discrete, and mixed data. The package is implemented the recent improvements in the Bayesian graphical models literature, including Mohammadi and Wit (2015) < https://doi:10.1214/14-BA889> and Mohammadi et al. (2017) < https://doi:10.1111/rssc.12171>. To speed up the computations, the BDMCMC sampling algorithms are implemented in parallel using OpenMP in C++.


News

BDgraph NEWS

  • CHANGES IN VERSION 2.19

    • The Title in Description is changed.
    • Function "I.g" is added.
  • CHANGES IN VERSION 2.20

    • Reversible jump MCMC algorithm is added to the "bdgraph()" fonction.
  • CHANGES IN VERSION 2.23

    • Function "I.g()" is chenged to "log_Ig()" and it is implemented in C++.
  • CHANGES IN VERSION 2.24

    • Function "phat()" is changed to "plinks()".
    • Function "prob()" is changed to "pgraph()".
    • Function "log_Ig()" is changed to "gnorm()".
  • CHANGES IN VERSION 2.28

    • The Title in Description is changed.
    • Functions "bdgraph.ts()", "rgcwish()", and "rcwish()" are added to the package.
  • CHANGES IN VERSION 2.36

    • The BDMCMC algorithms are implimented in parallel using OpenMP in C++.
  • CHANGES IN VERSION 2.38

    • Function "bdgraph.mpl()" is added to the package, which is based on Marginal Pseudo-Likelihood estimation.
    • In algorithm "gcgm", step "copula" are implimented in parallel using OpenMP in C++.
    • Function "bdgraph.ts" is implimented in parallel using OpenMP in C++.
    • New reference related to the ratio of normalizing constant is added to manual.
  • CHANGES IN VERSION 2.40

    • Option "g.prior" is added to function "bdgraph()", for prior distribution of the graph .
    • Option "cores" is added to function "bdgraph()", for determining the number of cores to use for parallel execution.
    • Function "transfer()" is added to the package.
  • CHANGES IN VERSION 2.41

    • Option "dgm-binary" is added to function "bdgraph()" which is for running Hill-Climbing algorihtm for binary data.

    • "configure" and "configure.ac" is added to configure checks of C++ codes.

  • CHANGES IN VERSION 2.42

    • Function "local_rates_ggm_mpl" in cpp is added for bdgraph.mpl function.
    • configure and configure.ac is added to the package for Makevars file.
    • Functions "bdgraph.ts", "rcwish", and "rcgwish" are removed from the package.
  • CHANGES IN VERSION 2.44

    • Functions "bdgraph.ts", "rcwish", and "rcgwish" are added to the package.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("BDgraph")

2.49 by Reza Mohammadi, 9 days ago


https://www.uva.nl/profile/a.mohammadi


Browse source code at https://github.com/cran/BDgraph


Authors: Reza Mohammadi [aut, cre] <https://orcid.org/0000-0001-9538-0648>, Ernst Wit [ctb], Adrian Dobra [ctb]


Documentation:   PDF Manual  


Task views: gRaphical Models in R, High-Performance and Parallel Computing with R


GPL (>= 2) license


Imports Matrix, igraph


Imported by bmixture, qgraph.

Depended on by ssgraph.


See at CRAN