Bayesian Factor Analysis

Provides model fitting for several Bayesian factor models including Gaussian, ordinal probit, mixed and semiparametric Gaussian copula factor models under a range of priors.


News

Version 0.4

  • Add accessor functions to extract (samples of) posterior loadings/scores
  • Remove references to the missing plot_loadings function in examples
  • Remove center.data and scale.data arguments to bfa_* functions
  • Doc fixes; also documenting the mean vector options in bfa_gauss and bfa_mixed

Version 0.3.3

  • Fixes some naming issues from C++11, per CRAN maintainers
  • Removes dependence on arma's RNG. Results from previous versions will not replicate exactly.

Version 0.3.2

  • bfa_gauss can include a mean vector by supplying additional (undocumented) arguments include_mean=TRUE and mean_mu0, mean_phi0 which are length p vectors giving the componentwise prior mean and precision (respectively) for the normal prior on E(y).

Version 0.3.1

  • Fixed segfault for print.status=0

  • Forcing CRAN binary to use later version of Rcpp, to get rid of the MatrixRow bug

Version 0.3

  • Fix for default MCMC simulation size of 0, which threw confusing errors.

  • Removed most plotting functions (the ones depending on ggplot2). The ggplot2 dependency was more trouble than it was worth. biplot still works, however.

  • Added a coef method, which returns posterior samples of the implied regression coefficients for conditional response distributions

  • Bugfix for imputing missing data.

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("bfa")

0.4 by Jared Murray, 2 years ago


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


Authors: Jared Murray


Documentation:   PDF Manual  


GPL-3 license


Imports coda, Rcpp

Linking to Rcpp, RcppArmadillo

System requirements: GNU make


See at CRAN