Bootstrap magnitude based inference

Calculates proportion of the estimator bootstrap resample lower/equivalent/higher distribution compared to SESOI band

bootstrap_MBI(bmbstats_object, estimator, SESOI_lower = 0, SESOI_upper = 0)

Arguments

bmbstats_object	Object of class bmbstats
estimator	Name of the estimator from the bmbstats_object
SESOI_lower	Lower smallest effect size of interest threshold
SESOI_upper	Upper smallest effect size of interest threshold

Value

bmbstats_MBI object

Examples

mean_MBI <- bootstrap_MBI(
  describe_data(rnorm(10, 100, 20)),
  estimator = "mean",
  SESOI_lower = 95,
  SESOI_upper = 105
)
#> Bootstraping: 2000 resamples
#> Calculating confidence intervals...
#> Warning: extreme order statistics used as endpoints
#> Warning: extreme order statistics used as endpoints
#> Warning: boot::boot.ci returned error or NULL when estimating CIs for min estimator. Returning NAs for upper and lower CIs
#> Done!
mean_MBI
#> Magnitude-based inference for the `mean` estimator
#> Bootstrap result: mean=98.768, 95% CI [91.44, 105.512]
#> SESOI: [95, 105]
#> 
#>        Test   prob
#>       lower 0.1390
#>  equivalent 0.8225
#>      higher 0.0385
#> 
#> Final inference: Likely equivalent
plot(mean_MBI, control = plot_control(points_jitter = FALSE))
#> Picking joint bandwidth of 0.689