Bug in equivalente_test when using big numbers for lower bounds

[DominiqueMakowski]

For some reason the equivalent tests says Accepted when the low bound becomes "big" (?)

m <- lm(Sepal.Length ~ Species, data=iris)
rez <- parameters::equivalence_test(m)
rez
#> # TOST-test for Practical Equivalence
#> 
#>   ROPE: [-0.08 0.08]
#> 
#> Parameter            |       90% CI |   SGPV | Equivalence |      p
#> -------------------------------------------------------------------
#> (Intercept)          | [4.89, 5.13] | < .001 |    Rejected | > .999
#> Species [versicolor] | [0.76, 1.10] | < .001 |    Rejected | > .999
#> Species [virginica]  | [1.41, 1.75] | < .001 |    Rejected | > .999

plot(rez)

rez <- parameters::equivalence_test(m, range=c(-Inf, 0.1))
rez
#> # TOST-test for Practical Equivalence
#> 
#>   ROPE: [-Inf 0.10]
#> 
#> Parameter            |       90% CI |   SGPV | Equivalence |      p
#> -------------------------------------------------------------------
#> (Intercept)          | [4.89, 5.13] | < .001 |    Accepted | < .001
#> Species [versicolor] | [0.76, 1.10] | < .001 |    Accepted | < .001
#> Species [virginica]  | [1.41, 1.75] | < .001 |    Accepted | < .001

plot(rez)

rez <- parameters::equivalence_test(m, range=c(-99, 0.1))
rez
#> # TOST-test for Practical Equivalence
#> 
#>   ROPE: [-99.00 0.10]
#> 
#> Parameter            |       90% CI |   SGPV | Equivalence |      p
#> -------------------------------------------------------------------
#> (Intercept)          | [4.89, 5.13] | < .001 |    Accepted | < .001
#> Species [versicolor] | [0.76, 1.10] | < .001 |    Accepted | < .001
#> Species [virginica]  | [1.41, 1.75] | < .001 |    Accepted | < .001

plot(rez)

^{Created on 2024-06-21 with reprex v2.1.0}

[strengejacke]

Yes, I think we were always assuming an "equal ranged" rope.