The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
|>
and thus R >= 4.1.id
variable identifying clustered observations
within the same individual a global rifttable()
option for
the entire data set, not only for specific estimators.type = "sum"
estimator.character
.design
accept weight
in addition
to weights
.event = "event_variable@Event_Type_One"
in the
design
. If multiple event types are present, estimate
cumulative incidence and differences/ratios of cumulative incidence in a
competing-event setting.weights
argument in the design
. Weighted
estimates are currently available for many survival estimators:
type = "cuminc"
, "surv"
, their differences and
ratios (e.g., "cumincdiff"
), and "hr"
. This is
a breaking change for Cox models (type = "hr"
), where
providing weights
in the arguments
list now
generates an error.@
in factor levels for a
table1_design()
.rt_gt()
: Output knitr-formatted tables for
GitHub-flavored markdown also in Quarto .qmd
, similar to
.Rmd
.exposure_levels
to let user
control handling of missing exposure levels (NA
) or factors
with empty levels as the exposure.type = "geomean"
for geometric means.exposure
or trend
of
""
as missing, and stratum = ""
as no
subsetting by the effect_modifier
, instead of subsetting to
effect modifier being an empty string. Input check that a stratum must
be provided for joint models and strata are not empty.type
as "blank"
.type2
has empty
results for some cells or if only a trend
variable and no
exposure
is given.NA
outcome
variables. More input checks.0
/1
outcomes in
log-linear models for ratios of continuous variables.type = "hr"
) allow for
weights
, clustering, and robust
standard
errors.ratio_digits_decrease
: By default, decrease
number of decimal digits shown for ratios by 1 digit for ratios > 3
and by 2 digits for ratios > 10. Leads to rounded ratios and
confidence intervals of 1.23
, 3.4
, and
11
.rt_gt()
now indents the first column and applies
markdown formatting to it by default.NA
instead of 0
in unstratified tables with all-null outcome.type = "maxfu"
ignored digits
and
diff_digits
.exposure
(strata labels) and
arguments
in one table.exposure
is
""
, not just for NA
.rt_gt()
: suppress random id
of gt tables
to keep git diff slim..event
, .outcome
, etc.
available under their original names.breastcancer
dataset from the risks
package.to
is set to ", "
by default, instead of
"-"
for ratio variables and " to "
for
difference variablestype
variable, following a restructuring of all estimation functions with
greater flexibility.design$type
no longer accepts additional arguments,
such as time points. Supply list instead via
design$arguments
.design$nmin
now differentiates between counts of total
observations or outcomes, depending on estimator.table1_design()
: Generate design of a
descriptive “Table 1.”outcome
option "variable@level"
for
categorical variables that displays level
as a binary
outcome. Used by table1_design()
.trend
variable in the design
) without an
exposure
.rifttable(reference = ...)
: Label for the reference
category.design$ci
: Width of confidence intervals.design$na_rm
: Omitting observations with missing
outcome data.design$arguments
: Flexibly passing along any argument
to estimation functions.breastcancer()
dataset|>
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.