Breaking changes

• The deprecated argument ppd was removed.

• Some of the deprecated arguments in plot() were removed.

• Pooling functions now use the model's degrees of freedom to calculate the
  critical values for the confidence intervals.

Changes

• test = "slope" (or test = "trend") are aliases in test_predictions() for
  test = NULL with numeric predictors.

• predict_response() (and ggpredict(), ggemmeans() and ggeffect()) get
  an argument bias_correction, to correct for bias when back-transforming
  predictions for non-Gaussian mixed models.

General

• Added support for models of class glm_weightit, ordinal_weightit,
  multinom_weightit from package WeightIt.

Bug fixes

• Fixed issues for multivariate response models due changes in the last
  insight package updates.

• Fixed issue with swapped lower and higher confidence interval values for
  models with inverse-link.

• Fixed CRAN check issues due to breaking changes in the last marginaleffects
  update.

Breaking

• The deprecated argument ci.lvl in test_predictions() was removed.

General

• ggpredict() now supports models of class glmgee (package glmtoolbox).

• ggemmeans() gains arguments vcov_fun, vcov_type and vcov_args to
  specify the variance-covariance matrix for the marginal means, similar to
  what is already available in ggpredict() and ggaverage().

• When test = "contrast", the engine is automatically set to "emmeans"
  in test_predictions().

• ggaverage() (or predict_response(..., margin = "empirical")) now also
  supports following type options for zero-inflated models: "zi_prob",
  "zero_inflated" and "fixed".

• Support for zero-inflated models was massively improved in test_predictions(),
  which, for instance, now also supports scale = "zprob" to calculate contrasts
  for the zero-inflation probability for zero-inflated models from glmmTMB or
  pscl. Furthermore, when predictions for zero-inflation probabilities were
  calculated using pr <- predict_response(..., type = "zi_prob"), corresponding
  contrasts will be calculated with test_predictions(pr) automatically.
  Additionally, other types for models with zero-inflation component
  ("zero_inflated", "fixed") are supported as well.

• ggeffect() now warns the user about arguments that are supported by
  ggpredict() or ggemmeans(), but not by this function (e.g., vcov_fun).

• Improved accuracy of standard errors for test_predictions(..., engine = "ggeffects").

• The terms argument now also accepts the shortcut "percentile" (plus numeric
  value) to select a range of percentiles for continuous variables, e.g.
  terms = "x [percentile90]" to select a range of the 90% percentile.

Bug fixes

• Fixed issue with brms models with monotonic effects in formula (mo()).

• Fixed issue in vcov() for ggeffects objects, which could occur in rare
  situations when some of the predictors were character vectors.

• Fixed issue with calculation of standard errors when one of the focal term
  was a character vector.

• Fixed issue in plot() method with show_data = TRUE, where in certain
  situations the raw data points were not colored when groups were present.

• Fixed issue in plot() with too many data points when collapsing random
  effects groups.

General

• ggpredict() now works for models of class clm2 from package ordinal,
  however, confidence intervals are not yet supported for these models.

• ggeffect() now passes the latent argument for models with ordinal outcome
  down to effects::Effect(), to plot effects for ordinal models on the latent
  scale.

• When argument test in test_predictions() is "interaction",
  "consecutive", or a data frame, emmeans is automatically used as backend,
  as this is the relevant package that supports these argument types.

• format() (and hence, print()) for test_predictions() gains a
  combine_levels argument, to combine levels of the focal term in the output
  table.

• The engine argument in test_predictions() can now also be "ggeffects".
  However, this is currently work-in-progress and offers muss less options as the
  default engine, "marginaleffects". It can be faster in some cases, though,
  and works for comparing predicted random effects in mixed models.

• test_predictions() now automatically falls back to engines "emmeans" or
  "ggeffects" if the marginaleffects (or emmeans) package is not installed.

• predict_response(), test_predictions() and ggpredict() will warn the user
  when all focal terms are only included as random effects in the model and no
  appropriate type or margin is specified. This is to avoid meaningless
  results.

• plot() gets an n_rows argument, to define the number of rows for the
  panel alignment. This is useful when the number of facets is large and the
  default alignment is not optimal.

• The ppd argument for Bayesian models will be superseded by the interval
  argument, i.e. ppd = TRUE is equivalent to interval = "prediction" (and
  ppd = FALSE is equivalent to interval = "confidence").

• When back_transform = FALSE, and model has a transformed response variable,
  the plot() method for ggeffects objects now rescales the raw data points.
  This ensures that the raw data points are plotted on the same scale as the
  predicted values when show_data = TRUE.

• Minor revisions of documentation and vignettes, to improve readability and
  clarity.

• Several arguments have been deprecated and replaced by new argument names. A
  warning is printed when deprecated arguments are used. The deprecated arguments
  will be removed in a future release.

Bug fixes

• Fixed issue in print() for ggeffect() and models with ordinal outcome,
  where one column was too much in the output.

• Fixed issue in test_predictions() with wrong order of term labels when
  a focal term was a character vector.

• Fixed issue in ggpredict() with wbm models from package panelr.

• Fixed issue in ggemmeans() for glmmTMB models with zero-inflation, when
  terms included variables that were specified in the conditional, but not
  in the zero-inflation model formula.

• Fixed issue in ggpredict() for Stan models (from packages rstanarm and brms)
  where the ci_level argument was not correctly recognized.

• Fixed CRAN check issues due to latest marginaleffects update.

General

• Overhaul of the documentation (again), to provide more clarity about the
  terminology "adjusted predictions", "marginal means" and "marginal effects",
  and how to calculate each of these quantities using the ggeffects package.

• print_html() methods were updated to work with the latest release of
  tinytable.

• New print_md() method, to print the output as markdown table. This is useful
  inside RMarkdown or Quarto documents, where the output can be directly included.

New functions

• predict_response() as "generic" high-level function, which is a replacement
  for ggpredict(), ggemmeans() and ggaverage(). The new function is more
  clear about how the function marginalizes over non-focal terms. The margin
  argument can be used to specify how to marginalize over non-focal terms, i.e.
  which function internally is used to compute the marginal effects.

General

• The documentation was revised, to provide more clarity about what the package
  functions do and how to decide, which function or method to calculate marginal
  effects is the most appropriate.

• Improved calculation of prediction intervals for Poisson regression models.

• Improved handling of the vcov_fun argument. This argument now accepts an
  estimation type as string, e.g. vcov_fun = "HC0", which is then used to
  compute the variance-covariance matrix. Thus, it is no longer necessary to
  define both vcov_fun and vcov_type, if the variance-covariance matrix is
  covered by one of the pre-defined estimation types. See ?ggpredict for
  details.

• hypothesis_test() now also accepts the vcov_fun argument, and not only
  vcov. This ensures consistency with the vcov_fun argument in ggpredict().
  Furthermore, the information about the type of variance-covariance matrix
  is saved to the ggeffects object returned by ggpredict(), predict_response()
  etc., and if this information is available, it is automatically used in
  hypothesis_test() when a ggeffects object is passed to the function.

Bug fixes

• Fixed bug in wrong order of printed (sub-)tables for predictions.

• Fixed wrong table column name for confidence interval columns for other
  confidence levels than the default 95% in print() for ggeffects objects.

• Fixed issue with ggpredict() for models of class fixest when the cluster
  variable was numeric.

Breaking Changes

• The print() method has been revised. A format() method was added, which
  allows to format the output of ggpredict() (and ggeffect() etc.) for
  printing. The refactoring of the print() method makes the code base easier
  to maintain and it is easier to enhance the print-functionality. Now it is
  possible to create HTML tables as well, using print_html(). The style of the
  output has also slightly changed. By default, confidence intervals are no
  longer enclosed in parentheses. You can change this behaviour by passing the
  ci_brackets argument to print() (see examples), or permanently define custom
  parentheses or brackets with, e.g., options(ggeffects_ci_brackets = c("[", "]")).
  Additionally, there are new arguments to further control the output of the
  tables: collapse_ci can be used to collapse confidence intervals into a
  single column together with the predicted values. collapse_tables can be used
  to collapse multiple tables into a single table (only applies when there is
  more than one focal term). Again, these settings can be permanently defined
  via options (see ?print.ggeffects for details).

New functions

• print_html(), to print the output as HTML table. This method is available
  for objects from ggpredict() (and alike) as well as hypothesis_test().

General

• A new vignette was added, showing examples for the new print-functionality.

Bug fixes

• Fixed issue with ggpredict() for models of class vglm with multivariate
  responses.

General

• ggpredict() now supports models of class rqs from package quantreg.

• Fixed issues to be compatible with forthcoming update of emmeans.

New function

• ggaverage(), to compute average predicted values. This function is based on
  marginaleffects::avg_predictons().

• pool_comparisons(), to pool results from multiple calls to hypothesis_test(),
  e.g. with imputed data sets.

General

• Support for sdmTMB (sdmTMB) models.

• Improved support for the logistf package, including models flic() and
  flac().

• Confidence intervals for predictions from merMod models (package lme4)
  now use the standard errors returned by predict(..., se.fit = TRUE). This
  should not affect numerical results, but can be more robust for certain edge
  cases. Note that standard errors are only based on predict() when
  tpye = "fixed". For type = "random", standard errors are still based on
  the model's variance-covariance matrix, taking uncertainty from random
  effects into account.

• hypothesis_test() now suppports models from package parsnip.

• johnson_neyman() gains a p_adjust argument, to adjust p-values for multiple
  comparisons. Currently, only p_adjust = "esarey" (resp. p_adjust = "es") and
  p_adjust = "fdr" (resp. p_adjust = "bh") are supported.

Bug fixes

• ggpredict() now computes appropriate predicted probabilites for models
  of class rms::lrm() with ordinal outcome.

• Fixed issue in ggpredict() for type = "random" when sampling from random
  effects levels, where the levels were numeric characters with a pattern like
  "001", "002", etc.

• Fixed minor issue in plot.ggalleffects().

• ... arguments in ggpredict() are now passed down to the predict() method
  for mgcv::gam() models.

Breaking changes

• Some function arguments will be renamed, to achieve consistency across the
  package and across other packages where I'm involved in the development.
  This will be a soft transition, i.e. the old argument names will still work
  for some package updates.

Changes

• The typical argument now supports a mix of functions for different variable
  types at which numeric or categorical covariates (non-focal terms) are held
  constant.

• Clarification of how the re.form argument is set when using type = "random"
  resp. type = "fixed" in ggpredict().

• hypothesis_test() now returns the standard error of contrasts or pairwise
  comparisons as attribute standard_error. This can be used to compute the
  test-statistic, if required. In forthcoming updates, there will be methods
  for insight::get_statistic() and parameters::model_parameters() to include
  standard errors and test-statistics in the output.

• test_predictions() was added as an alias for hypothesis_test().

Bug fixes

• Fixed issue in hypothesis_test() for mixed models, which sometimes failed
  when random effects group variables were numeric, and not factors.