Description
Usage
Arguments
Details
Value
Examples
View source: R/multisimsum.R
multisimsum
is an extension of simsum()
that can handle multiple estimated parameters at once.
multisimsum
calls simsum()
internally, each estimands at once.
There is only one new argument that must be set when calling multisimsum
: par
, a string representing the column of data
that identifies the different estimands.
Additionally, with multisimsum
the argument true
can be a named vector, where names correspond to each estimand (see examples).
Otherwise, constant values (or values identified by a column in data
) will be utilised.
See vignette("Ecustominputs", package = "rsimsum")
for more details.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15  (
,
,
estvarname,
se = ,
true = ,
methodvar = ,
ref = ,
= ,
ci.limits = ,
= ,
= ,
x = ,
control = ()
)

data 
A data.frame in which variable names are interpreted.
It has to be in tidy format, e.g. each variable forms a column and each observation forms a row.

par 
The name of the variable containing the methods to compare.
Can be NULL .

estvarname 
The name of the variable containing the point estimates.

se 
The name of the variable containing the standard errors of the point estimates.

true 
The true value of the parameter; this is used in calculations of bias, coverage, and mean squared error and is required whenever these performance measures are requested.
true can be a numeric value or a string that identifies a column in data .
In the former setting, simsum will assume the same value for all replications; conversely, each replication will use a distinct value for true as identified by each row of data .
See vignette("Ecustominputs", package = "rsimsum") for more details.

methodvar 
The name of the variable containing the methods to compare.
For instance, methods could be the models compared within a simulation study.
Can be NULL .
If a vector of column names is passed to simsum() , those columns will be combined into a single column named :methodvar using the base::interaction() function before computing all performance measures.

ref 
Specifies the reference method against which relative precision will be calculated.
Only useful if methodvar is specified.

by 
A vector of variable names to compute performance measures by a list of factors. Factors listed here are the (potentially several) datagenerating mechanisms used to simulate data under different scenarios (e.g. sample size, true distribution of a variable, etc.). Can be NULL .

ci.limits 
Can be used to specify the limits (lower and upper) of confidence intervals used to calculate coverage and biaseliminated coverage.
Useful for nonWald type estimators (e.g. bootstrap).
Defaults to NULL , where Waldtype confidence intervals based on the provided SEs are calculated for coverage; otherwise, it can be a numeric vector (for fixed confidence intervals) or a vector of strings that identify columns in data with replicationspecific lower and upper limits.
See vignette("Ecustominputs", package = "rsimsum") for more details.

df 
Can be used to specify that a column containing the replicationspecific number of degrees of freedom that will be used to calculate confidence intervals for coverage (and biaseliminated coverage) assuming tdistributed critical values (rather than normal theory intervals).
See vignette("Ecustominputs", package = "rsimsum") for more details.

dropbig 
Specifies that point estimates or standard errors beyond the maximum acceptable values should be dropped. Defaults to FALSE .

x 
Set to TRUE to include the data argument used to calculate summary statistics (i.e. after preprocessing the input dataset e.g. removing values deemed too large via the dropbig argument) as a slot. Calling simsum with x = TRUE is required to produce zipper plots. The downside is that the size of the returned object increases considerably, therefore it is set to FALSE by default.

control 
A list of parameters that control the behaviour of simsum . Possible values are:

mcse , whether to calculate Monte Carlo standard errors. Defaults to TRUE ;

level , the significance level used for coverage, biaseliminated coverage, and power. Defaults to 0.95;

power_df , whether to use robust critical values from a t distribution with power_df degrees of freedom when calculating power. Defaults to NULL , in which case a Gaussian distribution is used;

na.rm , whether to remove point estimates or standard errors where either (or both) is missing. Defaults to TRUE ;

char.sep , a character utilised when splitting the input dataset data . Generally, this should not be changed;

dropbig.max , specifies the maximum acceptable absolute value of the point estimates, after standardisation. Defaults to 10;

dropbig.semax , specifies the maximum acceptable absolute value of the standard error, after standardisation. Defaults to 100

dropbig.robust , specifies whether to use robust standardisation (using median and interquartile range) rather than normal standardisation (using mean and standard deviation). Defaults to TRUE , in which case robust standardisation will be used for dropbig .

The following names are not allowed for estvarname
, se
, methodvar
, by
, par
: stat
, est
, mcse
, lower
, upper
, :methodvar
.
An object of class multisimsum
.
 ("frailty", package = "rsimsum")
ms < (
= ,
= "par", true = (trt = 0.50, fv = 0.75),
estvarname = "b", se = "se", methodvar = "model",
= "fv_dist"
)
ms

rsimsum documentation built on May 21, 2021, 5:06 p.m.