Fix wrong computations of predictions for
arm::bayesglm() models.
Fix CRAN check issues.
Speed improvement for some models when calculating uncertainty intervals of predictions.
Minor fixes.
mo() with numeric predictors, which only allow to predict
for values that are actually present in the data.Fixed issue with adding raw data points for plots from logistic regression models, when the response variable was no factor with numeric levels.
Fixed issues with CRAN checks.
orm (package rms)Prediction intervals (where possible, or when
type = "random"), are now always based on sigma^2
(i.e. insight::get_sigma(model)^2). This is in line with
interval = "prediction" for lm, or for predictions
based on simulations (when type = "simulate").
print() now uses the name of the focal variable as
column name (instead) of "x").
collapse_by_group(), to generate a data frame where the
response value of the raw data is averaged over the levels of a (random
effect) grouping factor.A new vignette was added related to the definition and meaning of “marginal effects” and “adjusted predictions”. To be more strict and to avoid confusion with the term “marginal effect”, which meaning may vary across fields, either “marginal effects” was replaced by “adjusted predictions”, or “adjusted predictions” was added as term throughout the package’s documentation and vignettes.
Allow confidence intervals when predictions are conditioned on
random effect groups (i.e. when type = "random" and
terms includes a random effect group factor).
Predicted response values based on simulate()
(i.e. when type = "simulate") is now possible for more
model classes (see ?ggpredict).
ggpredict() now computes confidence intervals for
some edge cases where it previously failed (e.g. some models that do not
compute standard errors for predictions, and where a factor was included
in the model and not the focal term).
plot() gains a collapse.group argument,
which - in conjunction with add.data - averages
(“collapses”) the raw data by the levels of the group factors (random
effects).
data_grid() was added as more common alias for
new_data().
ggpredict() and plot() for
survival-models now always start with time = 1.
Fixed issue in print() for survival-models.
Fixed issue with type = "simulate" for
glmmTMB models.
Fixed issue with gamlss models that had
random() function in the model formula.
Fixed issue with incorrect back-transformation of predictions for
geeglm models.
residuals.type argument in plot() is
deprecated. Always using "working" residuals.pretty_range() and values_at() can now
also be used as function factories.
plot() gains a limit.range argument, to
limit the range of the prediction bands to the range of the
data.
Fixed issue with unnecessary back-transformation of log-transformed offset-terms from glmmTMB models.
Fixed issues with plotting raw data when predictor on x-axis was a character vector.
Fixed issues from CRAN checks.
interval to ggemmeans(), to
either compute confidence or prediction intervals.averaging (package MuMIn)pool_predictions(), to pool multiple
ggeffects objects. This can be used when predicted values
or estimated marginal means are calculated for models fit to multiple
imputed datasets.residualize_over_grid() is now
exported.log1p() and
log(mu + x).type = "random" or
"zi_random"), but random effects variances could not be
calculated or were almost zero.multinom
models in ggemmeans().ggemmeans() for models from
nlme.plot() for some models in
ggeffect().terms = "predictor [exp]" is no longer necessary.mlogit (package mlogit)plot() now can also create partial residuals plots.
There, arguments residuals, residuals.type and
residuals.line were added to add partial residuals, the
type of residuals and a possible loess-fit regression line for the
residual data.glm since
some time. Should be fixed now.ggpredict() and rlmerMods
models when using factors as adjusted terms.mclogit (package mclogit)ggeffect().ggpredict() gets a new type-option,
"zi.prob", to predict the zero-inflation probability (for
models from pscl, glmmTMB and
GLMMadaptive).add.data = TRUE in plot(), the raw data points
are also transformed accordingly.plot() with add.data = TRUE first adds the
layer with raw data, then the points / lines for the marginal effects,
so raw data points to not overlay the predicted values.terms-argument now also accepts the name of a
variable to define specific values. See vignette Marginal Effects at
Specific Values.vcov.type was not specified.type-argument.1.offset() terms.mixor (package mixor),
cgam, cgamm (package
cgam)x.as.factor is considered as less useful
and was removed.fixest (package fixest),
glmx (package glmx).plot(rawdata = TRUE) now also works for objects from
ggemmeans().ggpredict() now computes confidence intervals for
predictions from geeglm models.trials() as response
variable, ggpredict() used to choose the median value of
trials were the response was hold constant. Now, you can use the
condition-argument to hold the number of trials constant at
different values.print().clmm-models, when group factor in
random effects was numeric.emm() is discouraged, and so it was
removed.bracl, brmultinom (package
brglm2) and models from packages
bamlss and R2BayesX.plot() now uses dodge-position for raw data for
categorical x-axis, to align raw data points with points and error bars
geoms from predictions.show_pals()).vcov() function to calculate
variance-covariance matrix for marginal effects.ggemmeans() now also accepts type = "re"
and type = "re.zi", to add random effects variances to
prediction intervals for mixed models.... is now passed down to the
predict()-method for gamlss-objects, so
predictions can be computed for sigma, nu and tau as well.ggeffect(), when one term was a character vector.ggaverage() is discouraged, and so it was
removed.rprs_values() is now deprecated, the function
is named values_at(), and its alias is
representative_values().x.as.factor-argument defaults to
TRUE.ggpredict() now supports cumulative link and ordinal
vglm models from package VGAM.terms included random effects.add.data is an alias for the
rawdata-argument in plot().ggpredict() and ggemmeans() now also
support predictions for gam models from ziplss
family.print()-method for ordinal or cumulative link
models.plot()-method no longer changes the order of factor
levels for groups and facets.pretty_data() gets a length() argument to
define the length of intervals to be returned.values_at() is an alias for
rprs_values().betabin, negbin (package
aod), wbm (package panelr)ggpredict() now supports prediction intervals for
models from MCMCglmm.ggpredict() gets a
back.transform-argument, to tranform predicted values from
log-transformed responses back to their original scale (the default
behaviour), or to allow predictions to remain on log-scale (new).ggpredict() and ggemmeans() now can
calculate marginal effects for specific values from up to three terms
(i.e. terms can be of lenght four now).ci.style-argument from plot() now also
applies to error bars for categorical variables on the x-axis.gamlss, geeglm (package
geepack), lmrob and glmrob
(package robustbase), ols (package
rms), rlmer (package
robustlmm), rq and rqss
(package quantreg), tobit (package
AER), survreg (package
survival)terms = "predictor [1:10]") can now be changed with
by, e.g. terms = "predictor [1:10 by=.5]" (see
also vignette Marginal Effects at Specific Values).vcov.fun in ggpredict()) now also works for
following model-objects: coxph, plm,
polr (and probably also lme and
gls, not tested yet).ggpredict() gets an interval-argument, to
compute prediction intervals instead of confidence intervals.plot.ggeffects() now allows different horizontal and
vertical jittering for rawdata when jitter is
a numeric vector of length two.AsIs-conversion from division of two
variables as dependent variable, e.g. I(amount/frequency),
now should work.ggpredict() failed for MixMod-objects when
ci.lvl=NA.