bayesplot v1.15.0

• Add shape argument to mcmc_scatter by @behramulukir (#375)
• Restore pre ggplot2 v4.0 behavior of bayesplot theme setting by @jgabry (#385)
• Update loo PIT visual tests by @jgabry (#374)
• Update SVGs for visual tests for ggplot2 v4.0 by @jgabry (#379)
• Bumped roxygen and fixed linking note by @VisruthSK (#382)
• New pkgdown theme by @VisruthSK in (#378)
• Testthat 3e by @VisruthSK (#397)
• Fix height/width attributes in Rd files for CRAN by @jgabry (#387)

New Contributors

• @dependabot[bot] made their first contribution in #393

Full Changelog: v1.14.0...v1.15.0

bayesplot v1.14.0

What's Changed

New quantile dot plots (ppc_dots(), ppd_dots()), more support for discrete data in existing plots, support for user specified averaging functions in the PPC "avg" functions (ppc_scatter_avg(), ppc_error_scatter_avg(), etc.), and support for plotting binned errors against an x variable.

• add quantile dot plot functions by @behramulukir in #357
• add fun_avg to ppc_avg functions by @tjmahr in #349
• add x as argument to ppc_error_binned by @behramulukir in #359
• combining ppc_error_scatter_avg_vs_x with ppc_error_scatter_avg(x = x) by @behramulukir in #367
• adding new discrete style to ppc_rootogram by @behramulukir in #362
• adding discrete argument to stat functions by @behramulukir in #369

New Contributors

• @behramulukir made their first contribution in #357

Full Changelog: v1.13.0...v1.14.0

bayesplot v1.13.0

What's Changed

• Add ppc_loo_pit_ecdf() by @TeemuSailynoja in #345
• Add possibility for left-truncation to ppc_km_overlay() and ppc_km_overlay_grouped() by @Sakuski in #347
• Give user control over extrapolation in ppc_km_overlay() and ppc_km_overlay_grouped() by @Sakuski in #353
• Allow passing moment_match via ... for loo functions by @n-kall in #351
• Skip some tests when missing Suggests by @MichaelChirico in #344
• Remove a test that will fail with next ggplot2 release by @jgabry in #356

New Contributors

• @Sakuski made their first contribution in #347
• @n-kall made their first contribution in #351

Full Changelog: v1.12.0...v1.13.0

bayesplot v1.12.0

What's Changed

• Expand checking workflows to more platforms by @andrjohns (#324)
• Skip tests depending on Suggested dependency rstantools if not installed by @MichaelChirico (#325)
• Skip tests depending on Suggested dependency gridExtra if not installed by @MichaelChirico (#326)
• Fix missing legends for unobserved levels in rhat and neff plots (#328)
• Document problems with ppc_stat with stat="mean" (#329)
• Ensure rank overlay plot starts at 0 even if not all bins present, thanks @sims1253 (#332)
• Compatibility with ggplot2 3.6.0 by @teunbrand (#337)
• Update to latest GitHub actions workflows (#338)
• Remove deprecated functions from available_ppc() (#340)
• Fix missing counts in ppc_bars_data() by @TeemuSailynoja, thanks @famuvie (#342)

New Contributors

• @MichaelChirico made their first contribution in #326
• @andrjohns made their first contribution in #324

Full Changelog: v1.11.1...v1.12.0

bayesplot v1.11.1

What's Changed

Just a patch release to fix a minor bug:

• Limit maximum number of evaluation points in ppc_pit_ecdf functions by default to 1000. by @TeemuSailynoja in #318

Full Changelog: v1.11.0...v1.11.1

bayesplot v1.11.0

What's Changed

• Update for new ggplot2 release by @teunbrand in #309
• add 'bins' argument to many histogram plots by @paul-buerkner in #300
• Follow ggplot2 updates on facet_grid() and facet_wrap() by @heavywatal in #305
• better ppc_loo_pit_qq plots by @avehtari in #307
• Check prob is numeric for intervals plots by @tony-stone in #299
• Add bins and breaks arguments to more histogram and hex plots by @heavywatal in #313
• Replace size argument with linewidth for geom_line and geom_ridgeline by @heavywatal in #314
• All LOO plots now accept psis_object argument by @jgabry in #311
• ppc_loo_pit_ecdf() and ppc_loo_pit_ecdf_grouped() now support discrete variables, and their default method for selecting the number of ECDF evaluation points has been updated. @TeemuSailynoja in #316

New Contributors

• @teunbrand made their first contribution in #309
• @heavywatal made their first contribution in #305
• @tony-stone made their first contribution in #299

Full Changelog: v1.10.0...v1.11.0

bayesplot v1.10.0

• New function mcmc_rank_ecdf() for rank ecdf plots with confidence bands for
  assessing if two or more chains sample the same distribution (#282,
  @TeemuSailynoja)
• New functions ppc_pit_ecdf(), ppc_pit_ecdf_grouped(), PIT ecdf plots with
  confidence bands to assess if y and yrep contain samples from the same
  distribution. (#282, @TeemuSailynoja)
• Several ppc and ppd functions now accept the new linewidth argument
  introduced in ggplot2 3.4.0: ppc_bars(), ppc_bars_grouped(),
  ppc_intervals(), ppc_intervals_grouped(), ppd_intervals(),
  ppd_intervals_grouped().
• Fix bug in how mcmc_pairs() detected hitting max_treedepth, thanks to @dmphillippo. (#281)
• Fix failing tests due to changes in ggplot2 3.4.0 (#289)

bayesplot v1.9.0

Lots of new features in this release!

• New module PPD (posterior/prior predictive distribution) with a lot of new
  plotting functions with ppd_ prefix. These functions plot draws from the prior
  or posterior predictive distributions (PPD) without comparing to observed data
  (i.e., no y argument). Because these are not "checks" against the observed
  data we use PPD instead of PPC. These plots are essentially the same as the
  corresponding PPC plots but without showing any observed data (e.g.,
  ppd_intervals() is like ppc_intervals() but without plotting y). See
  help("PPD-overview") for details. (#151, #222)

• All PPC categories now have one or more _data() functions that return the
  data frame used for plotting (#97, #222). Many of these have already been in
  previous releases, but the new ones in this release are:

  • ppc_bars_data()
  • ppc_error_data()
  • ppc_error_binnned_data()
  • ppc_scatter_data()
  • ppc_scatter_avg_data()
  • ppc_stat_data()

• Many functions gain an argument facet_args for controlling ggplot2 faceting
  (many other functions have had this argument for a long time).
  The ones that just now got the argument are:

  • ppc_scatter()
  • ppc_scatter_avg_grouped()
  • ppc_error_hist()
  • ppc_error_hist_grouped()
  • ppc_error_scatter()
  • ppc_error_binned()

• New plotting function ppc_km_overlay_grouped(), the grouped variant of
  ppc_km_overlay(). (#260, @fweber144)

• ppc_scatter(), ppc_scatter_avg(), and ppc_scatter_avg_grouped() gain an
  argument ref_line, which can be set to FALSE to turn off the x=y line
  drawn behind the scatterplot.

• ppc_ribbon() and ppc_ribbon_grouped() gain argument y_draw that specifies whether the observed y should be plotted using a point, line, or both. (#257, @charlesm93)

• mcmc_*() functions now support all draws formats from the posterior package. (#277, @Ozan147)

• mcmc_dens() and mcmc_dens_overlay() gain arguments for controlling the
  the density calculation. (#258)

• mcmc_hist() and mcmc_dens() gain argument alpha for controlling transparency. (#244)

• mcmc_areas() and mcmc_areas_ridges() gain an argument border_size for
  controlling the thickness of the ridgelines. (#224)

bayesplot v1.8.0

Bug fixes

• mcmc_areas() tries to use less vertical blank space. (#218, #230)

• Fix bug in color_scheme_view() minimal theme (#213).

• Fix error in mcmc_acf() for certain input types. (#244, #245, @hhau)

New features

• New plotting functions ppc_dens_overlay_grouped() and ppc_ecdf_overlay_grouped()
  for plotting density and cumulative distributions of the posterior predictive
  distribution (versus observed data) by group. (#212)

• New plotting function ppc_km_overlay() for outcome variables that are
  right-censored. Empirical CCDF estimates of yrep are compared with the
  Kaplan-Meier estimate of y. (#233, #234, @fweber144)

• ppc_loo_pit_overlay() now uses a boundary correction for an improved kernel
  density estimation. The new argument boundary_correction defaults to TRUE but
  can be set to FALSE to recover the old version of the plot. (#171, #235,
  @ecoronado92)

• CmdStanMCMC objects (from CmdStanR) can now be used with extractor
  functions nuts_params(), log_posterior(), rhat(), and
  neff_ratio(). (#227)

• On the y axis, ppc_loo_pit_qq(..., compare = "normal") now plots standard
  normal quantiles calculated from the PIT values (instead of the standardized
  PIT values). (#240, #243, @fweber144)

• mcmc_rank_overlay() gains argument facet_args. (#221, @hhau)

• For mcmc_intervals() the sizeof the points and interval lines can be set withmcmc_intervals(..., outer_size, inner_size, point_size)`. (#215, #228, #229)

bayesplot v1.7.2

Minor internal fixes to ensure compatibility with dplyr 1.0.0