Skip to content

ci.limits without se doesn't work #58

Description

@ellessenne

Reproducible example:

library(rsimsum)
data("tt", package = "rsimsum")
# Works:
simsum(
  data = tt,
  estvarname = "diff",
  true = -1,
  se = "se",
  ci.limits = c("conf.low", "conf.high"),
  methodvar = "method",
  by = "dgm"
) |>
  summary(stats = "cover")
#> 'ref' method was not specified, 1 set as the reference
#> Values are:
#>  Point Estimate (Monte Carlo Standard Error)
#> 
#> Coverage of nominal 95% confidence interval:
#>  dgm               1               2
#>    1 0.9400 (0.0106) 0.9400 (0.0106)
#>    2 0.8780 (0.0146) 0.9420 (0.0105)
#>    3 0.9380 (0.0108) 0.9500 (0.0097)
#>    4 0.9020 (0.0133) 0.9420 (0.0105)
# Doesn't:
simsum(
  data = tt,
  estvarname = "diff",
  true = -1,
  ci.limits = c("conf.low", "conf.high"),
  methodvar = "method",
  by = "dgm"
) |>
  summary(stats = "cover")
#> 'ref' method was not specified, 1 set as the reference
#> Values are:
#>  Point Estimate (Monte Carlo Standard Error)

Created on 2025-02-11 with reprex v2.1.1

Likely just some control flow that omits coverage if se = NULL even though ci.limits is not?

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions