Optimal confidence sets for testing average bioequivalence

Tseng, Yu-Ling

doi:10.1007/bf02595733

Cited by 5 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we have reviewed a variety of simultaneous confidence regions for application in multiparameter bioequivalence studies. Statistical methods for confidence sets around normal mean vectors have been published on several occasions during the past 3 decades, with different optimality criteria such as minimum expected volume or minimum expected effective length being proposed, but most of these papers focused on mathematical theory and disregarded practical issues. As a consequence, there are a number of methods that work well in academic scenarios, eg, asymptotically, or with known variance, or when the PK parameters are uncorrelated, but whose performance in more realistic settings has been largely unclear, and hence, these developments have not made any impact on the way PK data are evaluated in practice.…”

Section: Discussionmentioning

confidence: 99%

“…Tseng considered a confidence region that is centred about the origin θ = 0 and minimises the expected effective length

ℓ_{e f f} (C (X)) \equiv 2 \sup_{y \in C (X)} | | y | |,

rather than the expected volume, at the origin. Tseng's confidence set is ill‐conditioned in that it can be empty; however, this does not impair the validity of the associated test.…”

Section: Methodsmentioning

confidence: 99%

“…The conservatism can be explained with the fact that the confidence region is grossly inflated as θ moves away from θ 0 . And under some circumstances, Tseng's region can be an empty set, which adds to the oddity of its behaviour. The bootstrap approach appears to be liberal for small and a bit conservative for large sample sizes.…”

Section: Simulation Studymentioning

confidence: 99%

“…The TOST procedure itself is easy to extend by applying the intersection-union principle, 12 and other multivariate equivalence tests are available as well, 13,14 but there have been controversies about how to construct a simultaneous confidence region around the vector of estimated PK measures. A number of ideas have been put forward in the past two decades: Various authors suggested methods tailored to (bio-)equivalence problems, [15][16][17][18][19] but also approaches that were not specifically designed for application in pharmaceutical statistics 20,21 may prove useful. For 3 and more PK parameters, methods that involve shrinkage (see Casella and Hwang 22 for an overview) seem appealing due to Stein's paradox.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Simultaneous confidence regions for multivariate bioequivalence

Pallmann

Jaki

2017

Statistics in Medicine

View full text Add to dashboard Cite

Demonstrating bioequivalence of several pharmacokinetic (PK) parameters, such as AUC and C , that are calculated from the same biological sample measurements is in fact a multivariate problem, even though this is neglected by most practitioners and regulatory bodies, who typically settle for separate univariate analyses. We believe, however, that a truly multivariate evaluation of all PK measures simultaneously is clearly more adequate. In this paper, we review methods to construct joint confidence regions around multivariate normal means and investigate their usefulness in simultaneous bioequivalence problems via simulation. Some of them work well for idealised scenarios but break down when faced with real-data challenges such as unknown variance and correlation among the PK parameters. We study the shapes of the confidence regions resulting from different methods, discuss how marginal simultaneous confidence intervals for the individual PK measures can be derived, and illustrate the application to data from a trial on ticlopidine hydrochloride. An R package is available.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Tseng considered a confidence region that is centred about the origin θ = 0 and minimises the expected effective length

ℓ_{e f f} (C (X)) \equiv 2 \sup_{y \in C (X)} | | y | |,

rather than the expected volume, at the origin. Tseng's confidence set is ill‐conditioned in that it can be empty; however, this does not impair the validity of the associated test.…”

Section: Methodsmentioning

confidence: 99%

Section: Simulation Studymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Simultaneous confidence regions for multivariate bioequivalence

Pallmann

Jaki

2017

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“…Below we use the Bonferroni inequality to develop a new method for family-wise error rate control for equivalence testing using TOST and illustrate that a naive application of the standard Bonferroni correction is unnecessarily conservative. Though relevant research in multiple comparisons (see Giani and Strassburger, 2000; Bofinger, 1985; Giani and Strassburger, 1994; Hsu, 1996; Tseng, 2002; Giani and Finner, 1991; Wellek, 2003) may produce more optimal solutions, our interest lies in simple rules that are easily motivated and implemented.…”

Section: Family-wise Error Rates For All Pair-wise Comparisonsmentioning

confidence: 99%

Easy Multiplicity Control in Equivalence Testing Using Two One-Sided Tests

Lauzon

Caffo

2009

The American Statistician

View full text Add to dashboard Cite

Equivalence testing is growing in use in scientific research outside of its traditional role in the drug approval process. Largely due to its ease of use and recommendation from the United States Food and Drug Administration guidance, the most common statistical method for testing equivalence is the two one-sided tests procedure (TOST). Like classical point-null hypothesis testing, TOST is subject to multiplicity concerns as more comparisons are made. In this manuscript, a condition that bounds the family-wise error rate using TOST is given. This condition then leads to a simple solution for controlling the family-wise error rate. Specifically, we demonstrate that if all pair-wise comparisons of k independent groups are being evaluated for equivalence, then simply scaling the nominal Type I error rate down by (k − 1) is sufficient to maintain the family-wise error rate at the desired value or less. The resulting rule is much less conservative than the equally simple Bonferroni correction. An example of equivalence testing in a non drug-development setting is given.

show abstract

Operability Region Equivalence: Simultaneous Confidence Bands for the Equivalence of Two Regression Models over Restricted Regions

2007

View full text Add to dashboard Cite

In many scientific problems the purpose of comparing two linear regression models is to demonstrate that they have only negligible differences and so can be regarded as being practically equivalent. The frequently used statistical approach of testing the homogeneity null hypothesis of the two models by using a partial F test is not appropriate for this purpose. In this paper, a simultaneous confidence band is proposed which provides an upper bound on the largest possible difference between the two models, in units of the standard error of the observations, over a given region of the covariates. This is demonstrated to be a more practical method for assessing the equivalence of the two regression models.

show abstract

Optimal confidence sets for testing average bioequivalence

Abstract: Average bioequivalence, Confidence set, Effective length, Multivariate bioequivalence testing, Optimal bioequivalence test, 62F03, 62F25,

Cited by 5 publications

References 15 publications

Simultaneous confidence regions for multivariate bioequivalence

Simultaneous confidence regions for multivariate bioequivalence

Easy Multiplicity Control in Equivalence Testing Using Two One-Sided Tests

Operability Region Equivalence: Simultaneous Confidence Bands for the Equivalence of Two Regression Models over Restricted Regions

Contact Info

Product

Resources

About