Skip to contents

The alpha-TOST Corrective Procedure for (Bio)Equivalence Testing

Source: R/internal_average_univariate.R
atost.Rd

Computes the alpha-TOST, a corrective procedure for the significance level applied to the Two One-Sided Test (TOST) for (bio)equivalence testing in the univariate framework.

Usage

atost(theta, sigma, nu, alpha, delta)

Arguments

theta

A numeric value corresponding to the estimated parameter of interest (such as a difference of means).

sigma

A numeric value corresponding to the estimated standard error of theta.

nu

A numeric value specifying the degrees of freedom.

alpha

A numeric value specifying the significance level.

delta

A numeric value defining the (bio)equivalence margin. The procedure assumes symmetry, i.e., the (bio)equivalence region is \((-\delta, \delta)\).

Value

An object of class tost with the following elements:

  • decision: Logical; indicates whether (bio)equivalence is accepted.

  • ci: Confidence region at the \(1 - 2\alpha\) level.

  • theta: The estimated difference(s) used in the test.

  • sigma: The estimated standard error used in the test.

  • nu: The degrees of freedom used in the test.

  • alpha: The significance level used in the test.

  • corrected_alpha: The significance level after adjustment.

  • delta: The (bio)equivalence limits used in the test.

  • method: The method used in the test ("alpha-TOST").

Examples

data(skin)

theta_hat = diff(apply(skin, 2, mean))
nu = nrow(skin) - 1
sig_hat = sd(apply(skin, 1, diff)) / sqrt(nu)
res_atost = cTOST:::atost(theta = theta_hat, sigma = sig_hat, nu = nu,
              alpha = 0.05, delta = log(1.25))
#> Error in alphahat.fun(sigma = sigma, nu = nu, alpha = alpha, delta = delta): could not find function "alphahat.fun"
compare_to_tost(res_atost)
#> Error: object 'res_atost' not found