Bayesian Synthetic Likelihood

Bayesian synthetic likelihood (BSL, Price et al. (2018) ) is an alternative to standard, non-parametric approximate Bayesian computation (ABC). BSL assumes a multivariate normal distribution for the summary statistic likelihood and it is suitable when the distribution of the model summary statistics is sufficiently regular. This package provides a Metropolis Hastings Markov chain Monte Carlo implementation of three methods (BSL, uBSL and semiBSL) and two shrinkage estimations (graphical lasso and Warton's estimation). uBSL (Price et al. (2018) ) uses an unbiased estimator to the normal density. A semi-parametric version of BSL (semiBSL, An et al. (2018) ) is more robust to non-normal summary statistics. Shrinkage estimations can help to bring down the number of simulations when the dimension of the summary statistic is high (e.g., BSLasso, An et al. (2019) ). Extensions to this package are planned.


BSL 2.0.0

  • Second major release. Please note some arguments have been renamed.
  • Added a new method ("semiBSL") in the main function "bsl".
  • Now bsl returns an S4 class object instead of S3. The plot method gives an option for whether ggplot2 or R graphics (the default) will be used.
  • The simulation function and summary statistics function do not need to have a list as the second argument now.
  • The new bsl function uses "parallelArgs" instead of "parallel_packages", where users can specify all arguments supported by "foreach".
  • Code update for three examples: MA(2), multivariate G & K, and cell biology.

Reference manual

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


3.0.0 by Ziwen An, a year ago

Browse source code at

Authors: Ziwen An [aut, cre] , Leah F. South [aut] , Christopher C. Drovandi [aut]

Documentation:   PDF Manual  

GPL (>= 2) license

Imports glasso, ggplot2, MASS, mvtnorm, copula, graphics, gridExtra, foreach, coda, Rcpp, methods

Suggests elliplot, doParallel

Linking to Rcpp, RcppArmadillo

See at CRAN