Release of bayesplot v1.16.0

Announcements
1

Version 1.16.0 of the bayesplot R package is now available! This release has quite a lot of new functionality and other improvements. Below are some highlights. For more detailed release notes that cover additional improvements and bug fixes see Changelog • bayesplot.

  • New ppc_calibration() plots (with many variants) check calibration for models with binary outcomes. There is also a new vignette with details and examples. Thanks to @TeemuSailynoja and @Florence_Bockting.

  • New method "correlated" for ppc_loo_pit_ecdf(), ppc_pit_ecdf(), and ppc_pit_ecdf_grouped(). This enables dependence-aware uniformity tests and visualization of influential ECDF regions. There is also a new vignette with details and examples. Thanks to @Florence_Bockting.

  • New quantile dot plots for MCMC draws: mcmc_dots() and mcmc_dots_by_chain(). Thanks to @behram.

  • New grouped rootogram plot via ppc_rootogram_grouped(). Thanks to @behram (and @mhollanders for the suggestion).

  • New show_marginal argument for ppd_*() distribution and test-statistic functions overlays the marginal predictive distribution on the plot. Thanks to @mattansb.

  • ppc_*() and ppd_*() functions now accept posterior::draws objects for predictive draws. Thanks to Ishaan Arora.

  • Compatibility with ggplot2 4.0 has also been greatly improved thanks to @BjarkeHautop, Ishaan Arora, and Utkarsh Pawade.

There are many other new features, improvements, and bug fixes in the release notes.

Thank you to everyone who contributed!

2

Several Bayesian Workflow book case studies were already using the development version of bayesplot, but now work with CRAN version. Case studies using ppc_pit_ecdf(..., method="correlated") and ppc_calibration() are

  • Ch 18 Nabiximols
  • Ch 24 Roaches
  • Ch 17 Sleepstudy
  • Ch 23 World cup
  • Ch 28 Variable selection