require("emmeans") emm_options(opt.digits = TRUE) knitr::opts_chunk$set(fig.width = 4.5, class.output = "ro")
emmGrid
objectemmGrid
objectsemmGrid
emmGrid
object {#update}
Several internal settings are saved when functions like ref_grid()
, emmeans()
,
contrast()
, etc. are run. Those settings can be manipulated via the update()
method for emmGrid
s. To illustrate, consider the pigs
dataset and model yet again:
pigs.lm <- lm(log(conc) ~ source + factor(percent), data = pigs) pigs.emm <- emmeans(pigs.lm, "source") pigs.emm
We see confidence intervals but not tests, by default. This happens as a result
of internal settings in pigs.emm.s
that are passed to summary()
when the
object is displayed. If we are going to work with this object a lot, we might
want to change its internal settings rather than having to rely on explicitly
calling summary()
with several arguments. If so, just update the internal
settings to what is desired; for example:
pigs.emm.s <- update(pigs.emm, infer = c(TRUE, TRUE), null = log(35), calc = c(n = ".wgt.")) pigs.emm.s
Note that by adding of calc
, we have set a default to calculate and
display the sample size when the object is summarized.
See help("update.emmGrid")
for details on the keywords
that can be changed. Mostly, they are the same as the names of arguments
in the functions that construct these objects.
Of course, we can always get what we want via calls to test()
, confint()
or
summary()
with appropriate arguments. But the update()
function is more
useful in sophisticated manipulations of objects, or called implicitly via the
...
or options
argument in emmeans()
and other functions. Those options are passed
to update()
just before the object is returned. For example, we could have
done the above update within the emmeans()
call as follows (results are not shown because they are the same as before):
emmeans(pigs.lm, "source", infer = c(TRUE, TRUE), null = log(35), calc = c(n = ".wgt."))
Speaking of the options
argument, note that the default in emmeans()
is options = get_emm_option("emmeans")
. Let's see what that is:
get_emm_option("emmeans")
So, by default, confidence intervals, but not tests, are displayed
when the result is summarized. The reverse is true for results of
contrast()
(and also the default for pairs()
which calls contrast()
):
get_emm_option("contrast")
There are also defaults for a newly constructed reference grid:
get_emm_option("ref_grid")
The default is to display neither intervals nor tests when summarizing.
In addition, the flag is.new.rg
is set to TRUE
, and that is why one
sees a str()
listing rather than a summary as the default when the object
is simply shown by typing its name at the console.
The user may have other preferences. She may want to see both intervals
and tests whenever contrasts are produced; and perhaps she also wants to
always default to the response scale when transformations or links
are present. We can change the defaults by setting the corresponding options;
and that is done via the emm_options()
function:
emm_options(emmeans = list(type = "response"), contrast = list(infer = c(TRUE, TRUE)))
Now, new emmeans()
results and contrasts follow the new defaults:
pigs.anal.p <- emmeans(pigs.lm, consec ~ percent) pigs.anal.p
Observe that the contrasts "inherited" the type = "response"
default from
the EMMs.
NOTE: Setting the above options does not change how existing emmGrid
objects
are displayed; it only affects ones constructed in the future.
There is one more option -- summary
-- that overrides all other display
defaults for both existing and future objects. For example, specifying
emm_options(summary = list(infer = c(TRUE, TRUE)))
will result in both
intervals and tests being displayed, regardless of their internal defaults,
unless infer
is explicitly specified in a call to summary()
.
To temporarily revert to factory defaults in a single call to emmeans()
or
contrast()
or pairs()
, specify options = NULL
in the call. To reset
everything to factory defaults (which we do presently), null-out all of the
emmeans package options:
options(emmeans = NULL)
When an emmGrid
object is summarized and displayed, the factory default is to
display it with just enough digits as is justified by the standard errors or HPD
intervals of the estimates displayed. You may use the "opt.digits"
option to
change this. If it is TRUE
(the default), we display only enough digits as is
justified (but at least 3). If it is set to FALSE
, the number of digits is set
using the R system's default, getOption("digits")
; this is often much more
precision than is justified. To illustrate, here is the summary of pigs.emm
displayed without optimizing digits. Compare it with the first summary in this
vignette.
emm_options(opt.digits = FALSE) pigs.emm emm_options(opt.digits = TRUE) # revert to optimal digits
By the way, setting this option does
not round the calculated values computed by summary.emmGrid()
or saved in a
summary)emm
object; it simply controls the precision displayed by
print.summary_emm()
.
The options accessed by emm_options()
and get_emm_option()
are stored in a
list named emmeans
within R's options environment. Therefore, if you desire
options other than the defaults provided on a regular basis, this can be
easily arranged by specifying them in your startup script for R. For example,
if you want to default to Satterthwaite degrees of freedom for lmer
models,
and display confidence intervals rather than tests for contrasts,
your .Rprofile
file could contain the line
options(emmeans = list(lmer.df = "satterthwaite", contrast = list(infer = c(TRUE, FALSE))))
emmGrid
objects {#rbind}
Two or more emmGrid
objects may be combined using the rbind()
or +
methods. The most common reason (or perhaps the only good reason) to do this
is to combine EMMs or contrasts into one family for purposes of applying
a multiplicity adjustment to tests or intervals.
A user may want to combine the three pairwise comparisons of sources
with the three comparisons above of consecutive percents into a single family of six tests with a suitable
multiplicity adjustment. This is done quite simply:
rbind(pairs(pigs.emm.s), pigs.anal.p[[2]])
The default adjustment is "bonferroni"
; we could have specified something different via the adjust
argument. An equivalent way to combine emmGrid
s is via the addition
operator. Any options may be provided by update()
. Below, we combine the same
results into a family but ask for the "exact" multiplicity adjustment.
update(pigs.anal.p[[2]] + pairs(pigs.emm.s), adjust = "mvt")
Also evident in comparing these results is that settings are obtained from the first object combined. So in the second output, where they are combined in reverse order, we get both confidence intervals and tests, and transformation to the response scale.
To subset an emmGrid
object, just use the subscripting operator []
.
For instance,
pigs.emm[2:3]
Sometimes, users want to use the results of an analysis (say, an emmeans()
call)
in other computations. The summary()
method creates a summary_emm
object
that inherits from the data.frame
class; so one may use the variables therein
just as those in a data frame.
An emmGrid
object has its own internal structure and we can't directly access
the values we see displayed. If follow-up computations are needed, use
summary()
(or confint()
or test()
), creates a summary_emm
object which
inherits from data.frame
-- making it possible to access the values.
For illustration, let's add the widths of the confidence intervals in our example.
CIs <- confint(pigs.emm) CIs$CI.width <- with(CIs, upper.CL - lower.CL) CIs
By the way, the values stored internally are kept to full precision, more than is typically displayed:
CIs$emmean
If you want to display more digits, specify so using the print
method:
print(CIs, digits = 5)
Sometimes, users want to group levels of a factor into a smaller number of groups.
Those groups may then be, say, averaged separately and compared, or used as a
by
factor. The add_grouping()
function serves this purpose. The function
takes four arguments: the object, the name of the grouping factor to be created,
the name of the reference factor that is being grouped, and a vector of level
names of the grouping factor corresponding to levels of the reference factor.
Suppose for example that we want to distinguish animal and non-animal sources of
protein in the pigs
example:
pigs.emm.ss <- add_grouping(pigs.emm.s, "type", "source", c("animal", "vegetable", "animal")) str(pigs.emm.ss)
Note that the new object has a nesting structure (see more about this in the "messy-data" vignette), with the reference factor nested in the new grouping factor. Now we can obtain means and comparisons for each group
emmeans(pigs.emm.ss, pairwise ~ type)
emmGrid
{#relevel}
Sometimes it is desirable to re-label the rows of an emmGrid
, or cast it in terms of
other factor(s). This can be done via the levels
argument in update()
.
As an example, sometimes a fitted model has a treatment factor that comprises combinations of other factors. In subsequent analysis, we may well want to break it down into the
individual factors' contributions. Consider, for example, the warpbreaks
data provided with R.
We will define a single factor and fit a non homogeneous-variance model:
warp <- transform(warpbreaks, treat = interaction(wool, tension)) library(nlme) warp.gls <- gls(breaks ~ treat, weights = varIdent(form = ~ 1|treat), data = warp) ( warp.emm <- emmeans(warp.gls, "treat") )
But now we want to re-cast this emmGrid
into one that has separate factors for wool
and tension
. We can do this as follows:
warp.fac <- update(warp.emm, levels = list( wool = c("A", "B"), tension = c("L", "M", "H"))) str(warp.fac)
So now we can do various contrasts involving the separate factors:
contrast(warp.fac, "consec", by = "wool")
Note: When re-leveling to more than one factor, you have to be careful to anticipate
that the levels will be expanded using expand.grid()
: the first factor in the list
varies the fastest and the last varies the slowest. That was the case in our example,
but in others, it may not be. Had the levels of treat
been
ordered as A.L, A.M, A.H, B.L, B.M, B.H
, then we would have had to specify the levels
of tension
first and the levels of wool
second.
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.