ggeffects
Create Tidy Data Frames of Marginal Effects for 'ggplot' from Model Outputs
Compute marginal effects and adjusted predictions from statistical models and returns the result as tidy data frames. These data frames are ready to use with the 'ggplot2'-package. Effects and predictions can be calculated for many different models. Interaction terms, splines and polynomial terms are also supported. The main functions are ggpredict(), ggemmeans() and ggeffect(). There is a generic plot()-method to plot the results using 'ggplot2'.
Why marginal effects?
Results of regression models are typically presented as tables that are easy to understand. For more complex models that include interaction or quadratic / spline terms, tables with numbers are less helpful and difficult to interpret. In such cases, marginal effects are far easier to understand. In particular, the visualization of marginal effects allows to intuitively get the idea of how predictors and outcome are associated, even for complex models.
Aim of this package
ggeffects computes marginal effects at the mean or average marginal effects from statistical models and returns the result as tidy data frames. These data frames are ready to use with the ggplot2-package.
Documentation and Support
Please visit https://strengejacke.github.io/ggeffects/ for documentation and vignettes. In case you want to file an issue or contribute in another way to the package, please follow this guide. For questions about the functionality, you may either contact me via email or also file an issue.
ggeffects supports many different models and is easy to use
Marginal effects can be calculated for many different models. Currently supported model-objects are: betareg, bglmer, blmer, brglm, brmsfit, clm, clm2, clmm, coxph, gam (package mgcv), Gam (package gam), gamlss, gamm, gamm4, gee, geeglm, glm, glm.nb, glmer, glmer.nb, glmmTMB, glmmPQL, glmrob, glmRob, gls, hurdle, ivreg, lm, lm_robust, lme, lmer, lmrob, lmRob, logistf, lrm, MixMod, MCMCglmm, multinom, nlmer, ols, plm, polr, rlm, rlmer, rq, rqss, stanreg, survreg, svyglm, svyglm.nb, tobit, truncreg, vgam, zeroinfl and zerotrunc. Other models not listed here are passed to a generic predict-function and might work as well, or maybe with ggeffect() or ggemmeans(), which effectively do the same as ggpredict().
Interaction terms, splines and polynomial terms are also supported. The main functions are ggpredict(), ggemmeans() and ggeffect(). There is a generic plot()-method to plot the results using ggplot2.
Examples
The returned data frames always have the same, consistent structure and column names, so it's easy to create ggplot-plots without the need to re-write the function call. x and predicted are the values for the x- and y-axis. conf.low and conf.high could be used as ymin and ymax aesthetics for ribbons to add confidence bands to the plot. group can be used as grouping-aesthetics, or for faceting.
ggpredict() requires at least one, but not more than three terms specified in the terms-argument. Predicted values of the response, along the values of the first term are calucalted, optionally grouped by the other terms specified in terms.
data(efc)
fit <- lm(barthtot ~ c12hour + neg_c_7 + c161sex + c172code, data = efc)
ggpredict(fit, terms = "c12hour")
#>
#> # Predicted values of Total score BARTHEL INDEX
#> # x = average number of hours of care per week
#>
#> x predicted std.error conf.low conf.high
#> 0 75.444 1.116 73.257 77.630
#> 15 71.644 0.965 69.753 73.535
#> 35 66.578 0.851 64.911 68.245
#> 50 62.779 0.852 61.108 64.449
#> 70 57.713 0.970 55.811 59.614
#> 85 53.913 1.122 51.713 56.113
#> 100 50.113 1.309 47.547 52.680
#> 120 45.047 1.591 41.929 48.166
#> 135 41.248 1.817 37.686 44.810
#> 170 32.382 2.373 27.732 37.033
#>
#> Adjusted for:
#> * neg_c_7 = 11.84
#> * c161sex = 1.76
#> * c172code = 1.97
A possible call to ggplot could look like this:
library(ggplot2)
mydf <- ggpredict(fit, terms = "c12hour")
ggplot(mydf, aes(x, predicted)) +
geom_line() +
geom_ribbon(aes(ymin = conf.low, ymax = conf.high), alpha = .1)
However, there is also a plot()-method. This method uses convenient defaults, to easily create the most suitable plot for the marginal effects.
mydf <- ggpredict(fit, terms = "c12hour")
plot(mydf)
ggeffects has a plot()-method with some convenient defaults, which allows quickly creating ggplot-objects.
With three variables, predictions can be grouped and faceted.
ggpredict(fit, terms = c("c12hour", "c172code", "c161sex"))
#>
#> # Predicted values of Total score BARTHEL INDEX
#> # x = average number of hours of care per week
#>
#> # low level of education
#> # [1] Male
#> x predicted std.error conf.low conf.high
#> 0 73.954 2.347 69.354 78.554
#> 45 62.556 2.208 58.228 66.883
#> 85 52.424 2.310 47.896 56.951
#> 170 30.893 3.085 24.847 36.939
#>
#> # low level of education
#> # [2] Female
#> x predicted std.error conf.low conf.high
#> 0 74.996 1.831 71.406 78.585
#> 45 63.597 1.603 60.456 66.738
#> 85 53.465 1.702 50.130 56.800
#> 170 31.934 2.606 26.827 37.042
#>
#> # intermediate level of education
#> # [1] Male
#> x predicted std.error conf.low conf.high
#> 0 74.673 1.845 71.055 78.290
#> 45 63.274 1.730 59.883 66.665
#> 85 53.142 1.911 49.397 56.887
#> 170 31.611 2.872 25.982 37.241
#>
#> # intermediate level of education
#> # [2] Female
#> x predicted std.error conf.low conf.high
#> 0 75.714 1.225 73.313 78.115
#> 45 64.315 0.968 62.418 66.213
#> 85 54.183 1.209 51.815 56.552
#> 170 32.653 2.403 27.943 37.362
#>
#> # high level of education
#> # [1] Male
#> x predicted std.error conf.low conf.high
#> 0 75.391 2.220 71.040 79.741
#> 45 63.992 2.176 59.727 68.258
#> 85 53.860 2.364 49.226 58.494
#> 170 32.330 3.257 25.946 38.713
#>
#> # high level of education
#> # [2] Female
#> x predicted std.error conf.low conf.high
#> 0 76.432 1.809 72.887 79.977
#> 45 65.034 1.712 61.679 68.388
#> 85 54.902 1.910 51.158 58.646
#> 170 33.371 2.895 27.697 39.045
#>
#> Adjusted for:
#> * neg_c_7 = 11.84
mydf <- ggpredict(fit, terms = c("c12hour", "c172code", "c161sex"))
ggplot(mydf, aes(x = x, y = predicted, colour = group)) +
stat_smooth(method = "lm", se = FALSE) +
facet_wrap(~facet)
plot() works for this case, as well:
plot(mydf)
There are some more features, which are explained in more detail in the package-vignette.
Contributing to the package
Please follow this guide if you like to contribute to this package.
Installation
Latest development build
To install the latest development snapshot (see latest changes below), type following commands into the R console:
library(devtools)devtools::install_github("strengejacke/ggeffects")Please note the package dependencies when installing from GitHub. The GitHub version of this package may depend on latest GitHub versions of my other packages, so you may need to install those first, if you encounter any problems. Here's the order for installing packages from GitHub:
sjlabelled → sjmisc → sjstats → ggeffects → sjPlot
Officiale, stable release
To install the latest stable release from CRAN, type following command into the R console:
install.packages("ggeffects")Citation
In case you want / have to cite my package, please use citation('ggeffects') for citation information.
News
ggeffects 0.10.0
General
- Better support, including confidence intervals, for some of the already supported model types.
- New package-vignette Logistic Mixed Effects Model with Interaction Term.
New supported models
gamlss,geeglm(package geepack),lmrobandglmrob(package robustbase),ols(package rms),rlmer(package robustlmm),rqandrqss(package quantreg),tobit(package AER),survreg(package survival)
Changes to functions
- The steps for specifying a range of values (e.g.
terms = "predictor [1:10]") can now be changed withby, e.g.terms = "predictor [1:10 by=.5]"(see also vignette Marginal Effects at Specific Values). - Robust standard errors for predictions (see argument
vcov.funinggpredict()) now also works for following model-objects:coxph,plm,polr(and probably alsolmeandgls, not tested yet). ggpredict()gets aninterval-argument, to compute prediction intervals instead of confidence intervals.plot.ggeffects()now allows different horizontal and vertical jittering forrawdatawhenjitteris a numeric vector of length two.
Bug fixes
- Models with
AsIs-conversion from division of two variables as dependent variable, e.g.I(amount/frequency), now should work. ggpredict()failed forMixMod-objects whenci.lvl=NA.
ggeffects 0.9.0
General
- Minor revisions to docs and vignettes.
- Reduce package dependencies.
- Better support, including confidence intervals, for some of the already supported model types.
- New package-vignette Customize Plot Appearance.
New supported models
ggemmeans()now supportstype = "fe.zi"for glmmTMB-models, i.e. predicted values are conditioned on the fixed effects and the zero-inflation components of glmmTMB-models.ggpredict()now supports MCMCglmm, ivreg and MixMod (package GLMMadaptive) models.ggemmeans()now supports MCMCglmm and MixMod (package GLMMadaptive) models.ggpredict()now computes confidence intervals for gam models (package gam).
New functions
new_data(), to create a data frame from all combinations of predictor values. This data frame typically can be used for thenewdata-argument inpredict(), in case it is necessary to quickly create an own data frame for this argument.
Changes to functions
ggpredict()no longer stops when predicted values with confidence intervals for glmmTMB- and other zero-inflated models can't be computed withtype = "fe.zi", and only returns the predicted values without confidence intervals.- When
ggpredict()fails to compute confidence intervals, a more informative error message is given. plot()gets aconnect.lines-argument, to connect dots from plots with discrete x-axis.
Bug fixes
ggpredict()did not work with glmmTMB- and other zero-inflated models, whentype = "fe.zi"and model- or zero-inflation formula had a polynomial term that was held constant (i.e. not part of theterms-argument).- Confidence intervals for zero-inflated models and
type = "fe.zi"could not be computed when the model contained polynomial terms and a very long formula (issue withdeparse(), cutting off very long formulas). - The
plot()-method put different spacing between groups when a numeric factor was used along the x-axis, where the factor levels where non equal-spaced. - Minor fixes regarding calculation of predictions from some already supported models
- Fixed issues with multiple response models of class
lminggeffects(). - Fixed issues with encoding in help-files.
ggeffects 0.8.0
General
- Minor changes to meet forthcoming changes in purrr.
- For consistency reasons, both
type = "fe"andtype = "re"return population-level predictions for mixed effects models (lme4, glmmTMB). The difference is thattype = "re"also takes the random effect variances for prediction intervals into account. Predicted values at specific levels of random effect terms is described in the package-vignettes Marginal Effects for Random Effects Models and Marginal Effects at Specific Values. - Revised docs and vignettes.
- Give more informative warning for misspelled variable names in
terms-argument. - Added custom (pre-defined) color-palettes, that can be used with
plot(). Useshow_pals()to show all available palettes. - Use more appropriate calculation for confidence intervals of predictions for model with zero-inflation component.
New supported models
ggpredict()andggeffect()now support brms-models with additional response information (liketrial()).ggpredict()now supports Gam, glmmPQL, clmm, and zerotrunc-models.- All models supported by the emmeans should also work with the new
ggemmeans()-function. Since this function is quite new, there still might be some bugs, though.
New functions
ggemmeans()to compute marginal effects by callingemmeans::emmeans().theme_ggeffects(), which can be used withggplot2::theme_set()to set the ggeffects-theme as default plotting theme. This makes it easier to add further theme-modifications likesjPlot::legend_style()orsjPlot::font_size().
Changes to functions
- Added prediction-type based on simulations (
type = "sim") toggpredict(), currently for models of class glmmTMB and merMod. x.catis a new alias for the argumentx.as.factor.- The
plot()-method gets aci.style-argument, to define different styles for the confidence bands for numeric x-axis-terms. - The
print()-method gets ax.lab-argument to print value labels instead of numeric values ifxis categorical. emm()now also supports all prediction-types, likeggpredict().
Bug fixes
- Fixed issue where confidence intervals could not be computed for variables with very small values, that differ only after the second decimal part.
- Fixed issue with
ggeffect(), which did not work if data had variables with more that 8 digits (fractional part longer than 8 numbers). - Fixed issue with multivariate response models fitted with brms or rstanarm when argument
ppd = TRUE. - Fixed issue with glmmTMB-models for
type = "fe.zi", which could mess up the correct order of predicted values forx. - Fixed minor issue with glmmTMB-models for
type = "fe.zi"ortype = "re.zi", when first terms had the[all]-tag. - Fixed minor issue in the
print()-method for mixed effects models, when predictions were conditioned on all model terms and adjustment was only done for random effects (output-line "adjusted for"). - Fixed issue for mixed models, where confidence intervals were not completely calculated, if
termsincluded a factor andcontrastswere set to other values thancontr.treatment. - Fixed issue with messed up order of confidence intervals for
glm-object and heteroskedasticity-consistent covariance matrix estimation. - Fixed issue for glmmTMB-models, when variables in dispersion or zero-inflation formula did not appear in the fixed effects formula.
- The
condition-argument was not always considered for some model types when calculating confidence intervals for predicted values.
ggeffects 0.7.0
General
- Support for monotonic predictors in brms models (
mo()). - For generalized additive models, values for splines are no longer automatically prettified (which ensures smooth plots, without the need to use the
[all]tag, i.e.terms="... [all]"). - If splines or plolynomial terms are used, a message is printed to indicate that using the
[all]tag, i.e.terms="... [all]", will produce smoother plots. - The package-vignette Marginal Effects at Specific Values now has examples on how to get marginal effects for each group level of random effects in mixed models.
- Revised
print()-method, that - for larger data frames - only prints representative data rows. Use then-argument inside theprint()-method to force a specific number of rows to be printed.
Changes to functions
- Added an
n-tag for theterms-argument inggpredict()andggeffect(), to give more flexibility according to how many values are used for "prettifying" large value ranges. - Added a
sample-tag for theterms-argument inggpredict()andggeffect(), to pick a random sample of values for plotting. ggpredict()andggeffect()now also return the standard error of predictions, if available.- The
jitter-argument inplot()now also changes the amount of noise for plots of models with binary outcome (whenrawdata = TRUE).
Bug fixes
- Fix issue with proper calculation of random effect variances for glmmTMB models for
type="re"andtype="re.zi"in general, and also for models withar1random effects structure.
ggeffects 0.6.0
General
- Reduce package dependencies.
- Moved package effects from dependencies to suggested packages, due to the restrictive requirements (R >= 3.5).
- New
print()-method, with a nicer print of the returned data frame. This method replaces thesummary()-method, which was removed. ggeffect()now supportsclm2-models from the ordinal-package.ggpredict()has improved support forcoxph-models from the survival-package (survival probabilities, cumulative hazards).
Changes to functions
- The
type-argument inggpredict()now has additional options,type = "fe.zi"andtype = "re.zi", to explicitely condition zero-inflated (mixed) models on their zero-inflation component. - The
type-argument inggpredict()now has additional options,type = "surv"andtype = "cumhaz", to plot probabilities of survival or cumulative hazards fromcoxph-models. ggpredict()gets argumentsvcov.fun,vcov.typeandvcov.argsto calculate robust standard errors for confidence intervals of predicted values. These are based on the varioussandwich::vcov*()-functions, hence robust standard errors can be calculated for all models that are supported bysandwich::vcov*().- The
plot()-method gets two argumentsline.sizeanddot.size, to determine the size of the geoms. - The
ci-argument for theplot()-method now also accepts the character values"dash"and"dot"to plot dashed or dotted lines as confidence bands. - The
terms-argument inggpredict()andggeffect()may also be a formula, which is more convenient for typing, but less flexible than specifying the terms as character vector with specific options.
Bug fixes
- Fixed improper calculation of confidence intervals for hurdle- and zero-inflated models (from package pscl), which could exceed the range of plausible values (e.g. below zero for incidence rates).
- Fixed issues with calculation of confidence intervals for mixed models with polynomial terms.
ggeffects 0.5.0
General
- New vignette Different Output between Stata and ggeffects.
Changes to functions
ggpredict()now automatically back-transforms predictions to the response scale for model with log-transformed response.ggeffect()andggpredict()now automatically set numeric vectors with 10 or more unique values to representative values (seerprs_values()), if these are used as second or third value in theterms-argument (to represent a grouping structure).- Fix memory allocation issue in
ggeffect(). rprs_values()is now exported.- The
pretty-argument is deprecated, because prettifying values almost always makes sense - so this is done automatically. ggpredict()now supportsbrmsfit-objects with categorical-family.ggalleffect()has been removed.ggeffect()now plots effects for all model terms ifterms = NULL.gginteraction()andggpoly()have been removed, asggpredict()andggeffect()are more efficient and generic for plotting interaction or polynomial terms.
Bug fixes
- Fix issues with categorical or ordinal outcome models (
polr,clm,multinom) forggeffect(). - Fix issues with confidence intervals for mixed models with log-transformed response value.
- Fix issues with confidence intervals for generalized mixed models when response value was a rate or proportion created with
cbind()in model formula.