- Better support for multivariate-response-models from
*brms*. - Support for cumulative-link-models from
*brms*. `ggpredict()`

now supports linear multivariate response models, i.e.`lm()`

with multiple outcomes.

`ggpredict()`

gets a`pretty`

-argument to reduce and “prettify” the value range from variables in`terms`

for predictions. This applies to all variables in`terms`

with more than 25 unique values.

- Recognize negative binomial family from
`brmsfit`

-models.

`ggpredict()`

,`ggeffect()`

and`gginteraction()`

get a`x.as.factor`

-argument to preserve factor-class for the`x`

-column in the returned data frame.- The
`terms`

-argument now also allows the specification of a range of numeric values in square brackets, e.g.`terms = "age [30:50]"`

.

- Give proper warning that
`clm`

-models don’t support`full.data`

-argument. `emm()`

did not work properly for some random effects models.

- Use
`convert_case()`

from*sjlabelled*, in preparation for the latest*snakecase*-package update.

- Model weights are now correctly taken into account.

- Support for
`brmsfit`

-models from the*brms*-package. - Support for
`clm`

-models from the*ordinal*-package. - Support for
`multinom`

-models from the*nnet*-package. - Posterior predictive distributions (see argument
`ppd`

) now compute uncertainty intervals also for non-gaussian models. - Use functions from package
*sjstats*(link inverse, model frame etc.). - If the regression model used weights,
`ggpredict()`

now computes the weighted mean as typical value for predictors that are held constant. - Use select-helpers from package
*tidyselect*, instead of*dplyr*.

- New
`summary()`

function, to provide information on predictions by grouping variables, and on constant values from adjustments.

`plot()`

gets a`show.legend`

-argument to show or hide the legend of plots.

- Fixed issues with
`gam`

- and`vgam`

-models.

`plot()`

gets a`dot.alpha`

-argument, to specify a different alpha-values for data points when plotting raw data.`plot()`

gets a`jitter`

-argument, to add a small amount of random variation to the location of data points when plotting raw data.`plot()`

and getter-functions (like`get_title()`

or`get_x_labels()`

) get a`case`

-argument, to convert labels into any case, using the snakecase-package.- Confidence intervals are now also computed for
`hurdle`

,`zeroinfl`

,`truncreg`

and`betareg`

-models. Note, however, that due to some uncertainty, the intervals may not be “smooth”.

- Confidence intervals for generalized mixed effects models are now computed properly.
- Different levels for confidence intervals (argument
`ci.lvl`

) were not always recognized. - Fixed issues with
`glmmTMB`

-models. - Fixed issues with
`lme`

-models. - Fixed issue when plotting data returned from
`ggeffect()`

, if the term in question was categorical.

- Support for
`stanreg`

models (pkg*rstanarm*). - Fixed issue with latest tidyr-update on CRAN.

- Plotting raw data with
`plot()`

did not work for predictions at specific values (i.e. when certain levels of predictor where selected in square brackets). - Computing predictions for
`mermod`

-objects did not work when model had only one fixed effects term.

- Updated package imports and dependencies.
- Support for
`polr`

models (pkg*MASS*). - Support for
`hurdle`

and`zeroinfl`

models (pkg*pscl*). - Support for
`betareg`

models (pkg*betareg*). - Support for
`truncreg`

models (pkg*truncreg*). - Support for
`coxph`

models (pkg*survival*).

`emm()`

as convenient shortcut to compute the estimate marginal mean of the model’s response value.

`plot()`

gets a`use.theme`

-argument, to use the default*ggeffects*-theme, or to use the default*ggplot*-theme.

- Fixed issues with columns resp. column names for models that used special functions in formula (e.g.
`s()`

for`gam`

-models, or`bs()`

for splines). - Fixed issue for wrong legend values when grouping term was a non-labelled factor with non-ordered numeric levels.

`ggpredict()`

computes proper confidence intervals for*merMod*- and*lme*-objects.- Improved
`plot()`

-method, to better plot raw data.

- Confidence intervals were not properly calculated for glm’s.
- For
`plot()`

, argument`rawdata`

did not work for models with discrete binary response. - Fixed issues with models of class
`lme`

and`glmmTMB`

. - Fixed issues with model-classes that preserved NA-values in model-frame.

- initial release