dropbig: Identify replications with large point estimates, standard...

In ellessenne/rsimsum: Analysis of Simulation Studies Including Monte Carlo Error

Identify replications with large point estimates, standard errors

Description

dropbig is useful to identify replications with large point estimates or standard errors. Large values are defined as standardised values above a given threshold, as defined when calling dropbig. Regular standardisation using mean and standard deviation is implemented, as well as robust standardisation using median and inter-quartile range. Further to that, the standardisation process is stratified by data-generating mechanism if by factors are defined.

Usage

dropbig(
  data,
  estvarname,
  se = NULL,
  methodvar = NULL,
  by = NULL,
  max = 10,
  semax = 100,
  robust = TRUE
)

Arguments

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.
estvarname	The name of the variable containing the point estimates.
se	The name of the variable containing the standard errors of the point estimates.
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.
by	A vector of variable names to compute performance measures by a list of factors. Factors listed here are the (potentially several) data-generating mechanisms used to simulate data under different scenarios (e.g. sample size, true distribution of a variable, etc.). Can be NULL.
max	Specifies the maximum acceptable absolute value of the point estimates, after standardisation. Defaults to 10.
semax	Specifies the maximum acceptable absolute value of the standard error, after standardisation. Defaults to 100.
robust	Specifies whether to use robust standardisation (using median and inter-quartile range) rather than normal standardisation (using mean and standard deviation). Defaults to TRUE.

Value

The same data.frame given as input with an additional column named .dropbig identifying rows that are classified as large (.dropbig = TRUE) according to the specified criterion.

Examples

data("frailty", package = "rsimsum")
frailty2 <- subset(frailty, par == "fv")

# Using low values of max, semax for illustration purposes:
dropbig(
  data = frailty2, estvarname = "b", se = "se",
  methodvar = "model", by = "fv_dist", max = 2, semax = 2
)

# Using regular standardisation:
dropbig(
  data = frailty2, estvarname = "b", se = "se",
  methodvar = "model", by = "fv_dist", max = 2, semax = 2, robust = FALSE
)

ellessenne/rsimsum documentation built on March 10, 2024, 1:21 p.m.