A Shrinkage Estimator for Combination of Bioassays

Abstract

A shrinkage estimator and a maximum likelihood estimator are proposed in this paper for combination of bioassays. The shrinkage estimator is obtained in closed form which incorporates prior information just on the common log relative potency after the homogeneity test for combination of bioassays is accepted. It is a practical improvement over other estimators which require iterative procedure to obtain the estimator for the relative potency. A real data is also used to show the superiorities for the newly-proposed procedures.

Robustness of the shrinkage estimator for the relative potency in the combination of multivariate bioassays

“…With the studies of combination of bioassays, most results appear in Bennett (1962), Armitage (1970), Meisner et al (1986), William (1988), Chen et al (1999), Xiong and Chen (2007), and Chen (2007). Chen et al (1999) proposed a novel empirical Bayesian estimator (EBE) and a shrinkage Bayesian estimator (SBE) for the relative potency from the combination of several multivariate bioassays by incorporating prior information on the model parameters based on Jeffreys' rules.…”