Estimating Relative Potency Using Prior Information

Abstract

This paper exhibits a closed-form point and interval estimator of the log relative potency that incorporates prior information on the log relative potency in a symmetric parallel line bioassay. The point estimator turns out to be a weighted average of the usual estimator and the prior mean where the weights are determined by the prior variance.

Concomitant information in bioassay and semi-parametric estimation

“…This is the approach taken in for example, Kim et al [5,7], Cho et al [8] and Chen et al [9]. We will refer to this as unadjusted method or approach.…”

“…The situation of no concomitant information, = 0 in (3), is covered in References [5,7], while the parametric concomitant formulation, (6), is treated in Reference [1,Chapter 12]. We can, with a slight modiÿcation, also approach the problem of estimating the log relative potency parameter, , as the second stage of a two-stage problem in the semi-parametric setting as well.…”

A Shrinkage Estimator for Combination of Bioassays

“…The most recent results on this issue appear in Kim, Carter and Hubert [7] , Kim, Cho, Carter and Hubert [9] ; and Kim, Carter, Hubert and Hand [8] . In these papers, the concern was just on the log relative potency parameter.…”

Point and Interval Estimation in the Combination of Bioassay Results with Unequal Variances

“…Non-parametric estimation procedure is discussed by Bennet (1984). The Bayesian inference in one-sample problem is given in Kim et al (1991Kim et al ( , 1992Kim et al ( , 1993, and others.…”