Index of vignette topics • emmeans

Jump to: A B C D E F G H I J K L M N O P Q R S T U V W X Z

C

Causal inference
cld()
clm models
coda package
coef()
Cohen’s d
Compact letter displays
Comparison arrows
- Derivation
Comparisons
- Back-transforming
- Displaying as groups
- Displaying P values
- Graphical
  - How arrows are determined
- with logs
- Obtaining differences rather than ratios
- with overlapping CIs
Comparisons result in (nothing)
Confidence intervals
- Overlapping
confint()
Confounded effects
Confounding
consec contrasts
Constrained marginal means
Consultants
Containment d.f.
contrast()
- adjust
- Changing defaults
- combine
- interaction
- Linear functions
- simple
  - in confidence-intervals: simple
  - in interactions: simple
Contrasts
- of other contrasts
- Custom
- Formula
- Multivariate
- Pairwise
- Polynomial
- Tests of
  - with transformations
Controlled experiments
Count regression
Counterfactuals
cov.reduce
Covariates
- Adjusted
- cov.keep
- cov.reduce
- Derived
- emmeans() doesn’t work
- Interacting with factors
- Mediating
cross-group comparisons
cross.adjust

D

E

eff contrasts
eff_size()
Effect size
emm_basis()
- Arguments and returned value
- Communicating with recover_data()
- Dispatching
- Hook functions
- for lqs objects
- Re-gridded
- for rsm objects
emm_list object
emm_options()
.emmc functions
emmeans package
- Exporting extensions to
- Quick-start guide
emmeans()
- And the underlying model
- Arguments passed to ref_grid()
- Changing defaults
- Fastest way to get wrong answers
- Surprising results from
- weights
  - in basics: weights
  - in messy-data: weights
- With transformations
emmGrid objects
- Accessing data
- Combining and subsetting
- Modifying
- Setting defaults for
emmip()
- in basics: emmip
- in basics: plots
- in interactions: factors
- nested factors
EMMs
- Projecting to a submodel
- What are they?
emtrends()
- in interactions: covariates
- in interactions: oranges
estHook
Estimability
Estimability issues
Estimable functions
Estimated marginal means
- Definition of
- Demonstrated
- Manually computed
- Reference grid
- vs. ordinary marginal means
Examples
- auto.noise
- Bayesian model
- cbpp
  - in sophisticated: mcmc
  - in transformations: cbpp
- ChickWeight
- cows
- feedlot
- fiber
  - in interactions: covariates
  - in transformations: stdize
- framing
- Gamma regression
- InsectSprays
- Insurance
- Insurance claims (SAS)
- Logistic regression
- lqs objects
- MOats
- mtcars
  - in basics: depcovs
  - in messy-data: nuis.example
- Multivariate
- Nested fixed effects
- neuralgia
  - counterfactuals
  - in transformations: links
  - in transformations: trangraph
- nutrition
  - submodel
  - weights
- Oats
- oranges
  - in comparisons: formulas
  - in interactions: oranges
- Ordinal model
- pigs
  - in basics: pigs
  - in confidence-intervals: summary
  - in transformations: altscale
  - in transformations: overview
  - in transformations: pigs-biasadj
- rlm objects
- Robust regression
- Split-plot experiment
- Unbalanced data
- warpbreaks
  - in confidence-intervals: cross-adjust
  - in transformations: tranlink
  - in utilities: relevel
- Welch’s t comparisons
- wine
Expected marginal means
- Foundations of
- Reliance on a model
Experimental versus observational data
Exporting output
Extending emmeans
- Exports useful to developers
- Restrictions

F

F test
- vs. pairwise comparisons
- Role in post hoc tests
Factors
- Mediating
Formatting results
Foundations of EMMs
Frequently asked questions

G

H

I

J

K

L

Labels
- Changing
Large models
Latin squares
Least-squares means
Levels
- Changing
Linear functions
Link functions
lme models
lmerMod models
Logistic-like regression
- Non-logit links
Logistic regression
LSD
- protected

M

make.tran()
mcmc objects
Means
- Generalized
- Weighted
Mediating covariates
Memory usage
- Limiting
mira models
misc attribute and argument
misc$regrid.flag
Missing cells
mlm models
mmer models
mode argument
- implied re-gridding
Model
- Get it right first
- Importance of
  - in basics: models
  - in FAQs: fastest
- Importance of getting it right
Model averaging
Modeling
- pigs example
- Response transformation
Modelling
Models
- Constrained
- Large
- Quick reference
- Unsupported
Multi-factor studies
Multinomial models
Multiple imputation
Multiplicity adjustments
Multivariate contrasts
Multivariate models
- in basics: multiv
- in interactions: oranges
- in models: M
- with submodel
Multivariate t ("mvt") adjustment
mvcontrast()
mvtnorm package

N

O

Observational data
- in basics: weights
- in messy-data: issues
Observational versus experimental data
Odds ratios
- but I wanted differences
Offsets
- Models with offsets
- offset() term vs. offset argument in model
Only one mean
opt.digits option
Options
- Startup
Ordinal models

P

P values
- Adjusted
- Adjustment is ignored
- Interpreting
pairs()
pairwise ~ factors
- Why you should avoid this
Pairwise comparisons
- Matrix displays
pairwise contrasts
Pairwise P-value plots
params
Percentage differences
plot()
- nested factors
plot.emmGrid()
Plots
- of confidence intervals
- of EMMs
- Enhancing with ggplot2 functions
- Interaction-style
+ operator
Poisson regression
- Surprising results
polreg models
Polynomial regression
Pooled t
- Instead of Welch’s t
postGridHook
Practices, recommended
Precision
- Displaying results with more digits
Predictions
- Bayesian models
  - in predictions: bayes
  - in sophisticated: predict-mcmc
- Error SD
- graphics
- on Particular strata
- Posterior predictive distribution
- Reference grid
- Total SD
print.summary_emm()
Probit regression
- Odds or risk ratios
pwpm()
pwpp()

Q

R

R-squared
Random predictors
- Accessing levels
Random slopes
- Bias adjustment
Rank deficiency
Ratios
- but I wanted differences
rbind()
Re-gridding
- Implied
Re-labeling
Recommended practices
recover_data()
- Communicating with emm_basis()
- data and params arguments
- Dispatching
- Error handling
- for lqs objects
- for rsm objects
recover_data.call()
- frame argument
ref_grid()
- at
- cov.reduce
  - in basics: plot.emmGrid
  - in basics: RG
- mult.name
- nesting
- offset
- regrid argument
Reference grid
- Covariates
- Defined
- Indicator variables
- Predictions on
Reference grids
- Prediction on
Region of practical equivalence
Registering recover_data and emm_basis methods
regrid argument
regrid()
- probit or other lon-logit models
- regrid vs. type
- to obtain risk ratios
- transform = "log"
regrid.flag
Residual plots
Response scale
Response transformations
revpairwise contrasts
rg.limit option
Risk ratios
RMarkdown
ROPE
rsm package
rstanarm

S

Sample size, displaying
Sandwich estimators
Satterthwaite d.f.
- in models: K
- in models: L
"scale" type
scale()
Selecting results
Sidak adjustment
Significance
- Assessing
simple = "each"
Simple comparisons
- in confidence-intervals: simple
- in FAQs: interactions
- in interactions: simple
Simpson’s paradox
specs
- Formula
Standardized response
stanreg objects
* gazing (star gazing)
Startup options
Statistical consultants
Statistics is hard
str()
submodel
- "minimal" and "type2"
- in a multivariate model
- in messy-data: submodels
- in xplanations: submodels
Subsets of data
summary()
- adjust
- in basics: emmobj
- Bayesian models
  - in confidence-intervals: summary
  - in models: S
- Calculated columns
- in confidence-intervals: summary
- HPD intervals
- hpd.summary()
- infer
  - in comparisons: pairwise
  - in confidence-intervals: summary
- Show sample size
- type = "unlink"
summary_emm object
- As a data frame

U

V

W

X

xtable method

Z

Index generated by the vigindex package.