Abstract: Standard hypothesis tests are set up to prove differences from a standard or between populations. The goal in many applications is to demonstrate with statistical confidence that in fact there is no difference. Common examples include performance between model & simulation results versus live events, generic versus brand drug safety and efficacy, material engineering properties between different treatments, and many other practical problems. While a Test of Equivalence that uses the Two One-Sided Tests (TOST) is useful, in practice we often require more than equal means to properly characterize the similarity. We will demonstrate methods to help establish equivalence in means, variances, and distributions as an introduction. The focus will be on the equality between models; that is, not only the output responses being approximately equal, but also the consistency in the parameters characterizing the process.
Webinar Goals:
- Understand why failing to reject a one or two sample t-test null hypothesis does not constitute adequate proof
- Evaluate equivalence with TOST
- Evaluate equivalence for distributions and variance
- Know how to test if regression coefficients are equivalent
- Understand how two determine if two or more curves are equivalent with functional data analysis