![](../logo.png)
Index of vignette topics
emmeans package, Version 1.10.3
Source:vignettes/vignette-topics.Rmd
vignette-topics.Rmd
# {##}
- a probability scale, and whatever link was used (say, probit) has already been
- accounted for, so is not “remembered” for possible later back-transformation. In
add_grouping()
addl.vars
-
adjust
- Adjusted means
- Adjusted R-squared
afex_aov
objects- Alias matrix
- Analysis of subsets of data
- Analysis of variance
aovList
objectsappx-satterthwaite
method- Arguments
-
as.mcmc()
- ASA Statement on P values
- Asymptotic tests
- ATOM
averaging
models
[ {#[}
< {#<}
- Causal inference
cld()
clm
models- coda package
coef()
- Cohen’s d
- Compact letter displays
- Comparison arrows
- Comparisons
- Comparisons result in
(nothing)
- Confidence intervals
confint()
- Confounded effects
- Confounding
consec
contrasts- Constrained marginal means
- Consultants
- Containment d.f.
-
contrast()
- Contrasts
- Controlled experiments
- Count regression
- Counterfactuals
cov.reduce
- Covariates
- cross-group comparisons
cross.adjust
B
eff
contrastseff_size()
- Effect size
-
emm_basis()
emm_list
objectemm_options()
.emmc
functions- emmeans package
-
emmeans()
-
emmGrid
objects -
emmip()
- EMMs
-
emtrends()
estHook
- Estimability
- Estimability issues
- Estimable functions
- Estimated marginal means
- estimating marginal means will be averages of the probabilities in the reference
- Examples
auto.noise
- Bayesian model
-
cbpp
ChickWeight
cows
feedlot
-
fiber
framing
- Gamma regression
InsectSprays
Insurance
- Insurance claims (SAS)
- Logistic regression
lqs
objectsMOats
-
mtcars
- Multivariate
- Nested fixed effects
-
neuralgia
-
nutrition
Oats
-
oranges
- Ordinal model
-
pigs
rlm
objects- Robust regression
- Split-plot experiment
- Unbalanced data
-
warpbreaks
- Welch’s t comparisons
wine
- Expected marginal means
- Experimental versus observational data
- Exporting output
- Extending emmeans
C
D
- G-Computation
gam
modelsgamlss
models- GEE models
- Generalized additive models
- Generalized linear models
- Geometric means
- Get the model right first
get_emm_option()
- ggplot2 package
- GIGO (garbage in, garbage out)
glm
xxx modelsgls
models- Graphical displays
- grid
- Grouping factors
- Grouping into separate sets
I
- Labels
- Large models
- Latin squares
- Least-squares means
- Levels
- Linear functions
- Link functions
lme
models-
lmerMod
models - Logistic-like regression
- Logistic regression
- LSD
J
make.tran()
mcmc
objects- Means
- Mediating covariates
- Memory usage
mira
modelsmisc
attribute and argument- Missing cells
mlm
modelsmmer
models-
mode
argument - Model
- Model averaging
- Modeling
- Modelling
- Models
- models for ordinal data allow
for a
"prob"
mode that produces estimates of - Multi-factor studies
- Multinomial models
- Multiple imputation
- Multiplicity adjustments
- Multivariate contrasts
- Multivariate models
- Multivariate t
(
"mvt"
) adjustment mvcontrast()
- mvtnorm package
X
L
- Observational data
- Observational versus experimental data
- Odds ratios
- Offsets
- Only one mean
opt.digits
option- Options
- Ordinal models
M
- P values
pairs()
-
pairwise ~ factors
- Pairwise comparisons
pairwise
contrasts- Pairwise P-value plots
params
- Percentage differences
-
plot()
plot.emmGrid()
- Plots
+
operator- Poisson regression
polreg
models- Polynomial regression
- Pooled t
postGridHook
- Practices, recommended
- Precision
- Predictions
- preempting any timing choices you might otherwise have made about handling the
print.summary_emm()
- probabilities for each ordinal level. The reference grid comprises estimates on
- Probit regression
pwpm()
pwpp()
O
- R-squared
- Random predictors
- Random slopes
- Rank deficiency
- Ratios
rbind()
- Re-gridding
- Re-labeling
- Recommended practices
-
recover_data()
-
recover_data.call()
-
ref_grid()
- Reference grid
- Reference grids
- Region of practical equivalence
- Registering
recover_data
andemm_basis
methods regrid
argument-
regrid()
regrid()
that takes place at the time the reference grid is constructed,- Residual plots
- Response scale
- Response transformations
revpairwise
contrastsrg.limit
option- Risk ratios
- RMarkdown
- ROPE
- rsm package
rstanarm
P
- Sample size, displaying
- Sandwich estimators
- Satterthwaite d.f.
"scale"
typescale()
- Selecting results
- Sidak adjustment
- Significance
- Similar things happen with certain options with multinomial models,
simple = "each"
- Simple comparisons
- Simpson’s paradox
- so they will be different than what you would have obtained by keeping
- Some model classes provide
special argument(s) (typically
mode
) that may cause -
specs
- Standardized response
stanreg
objects- * gazing (star gazing)
- Startup options
- Statistical consultants
- Statistics is hard
str()
-
submodel
- Subsets of data
-
summary()
-
summary_emm
object
Q
- t tests vs. z tests
-
test()
- Tests
- that sense, when we use
mode = "prob"
, it is sort of like an implied call to - the link scale.
- things on the link scale and then computing the probabilities after averaging on
- Too few means
- transformation. If there are one or more factors that are averaged over in
- Transformations
- transformations or links to be handled early. For example, cumulative link
- Transformations|Implied re-gridding
- Transformations|Special modes
- Trends
trt.vs.ctrl
contrasts- Tukey adjustment
type
type = "response"
type = "scale"
- Type II analysis
- Type III tests
S
- Variables that are not predictors
vcov.
vcovHook
- Vignettes
W
- z tests
- zero-inflated, or hurdle models. Those special modes are a great convenience
zeroinfl
models
Index generated by the vigindex package.