simsum  R Documentation 
simsum()
computes performance measures for simulation studies in which each simulated data set yields point estimates by one or more analysis methods.
Bias, relative bias, empirical standard error and precision relative to a reference method can be computed for each method.
If, in addition, modelbased standard errors are available then simsum()
can compute the average modelbased standard error, the relative error in the modelbased standard error, the coverage of nominal confidence intervals, the coverage under the assumption that there is no bias (biaseliminated coverage), and the power to reject a null hypothesis.
Monte Carlo errors are available for all estimated quantities.
simsum(
data,
estvarname,
se = NULL,
true = NULL,
methodvar = NULL,
ref = NULL,
by = NULL,
ci.limits = NULL,
df = NULL,
dropbig = FALSE,
x = FALSE,
control = list()
)
data 
A 
estvarname 
The name of the variable containing the point estimates. Note that some column names are forbidden: these are listed below in the Details section. 
se 
The name of the variable containing the standard errors of the point estimates. Note that some column names are forbidden: these are listed below in the Details section. 
true 
The true value of the parameter; this is used in calculations of bias, relative bias, coverage, and mean squared error and is required whenever these performance measures are requested.

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 
ref 
Specifies the reference method against which relative precision will be calculated.
Only useful if 
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 
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 
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 
dropbig 
Specifies that point estimates or standard errors beyond the maximum acceptable values should be dropped. Defaults to 
x 
Set to 
control 
A list of parameters that control the behaviour of

The following names are not allowed for any column in data
that is passed to simsum()
: stat
, est
, mcse
, lower
, upper
, :methodvar
, :true
.
An object of class simsum
.
White, I.R. 2010. simsum: Analyses of simulation studies including Monte Carlo error. The Stata Journal 10(3): 369385. https://www.statajournal.com/article.html?article=st0200
Morris, T.P., White, I.R. and Crowther, M.J. 2019. Using simulation studies to evaluate statistical methods. Statistics in Medicine, \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1002/sim.8086")}
Gasparini, A. 2018. rsimsum: Summarise results from Monte Carlo simulation studies. Journal of Open Source Software 3(26):739, \Sexpr[results=rd]{tools:::Rd_expr_doi("10.21105/joss.00739")}
data("MIsim", package = "rsimsum")
s < simsum(data = MIsim, estvarname = "b", true = 0.5, se = "se", methodvar = "method", ref = "CC")
# If 'ref' is not specified, the reference method is inferred
s < simsum(data = MIsim, estvarname = "b", true = 0.5, se = "se", methodvar = "method")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.