Stein phenomenon Summary Several procedures designed to reduce nonconformity in interlaboratory studies by shrinking data toward a consensus matrix weighted mean are suggested. Some of them are shown to have a smaller quadratic risk than the vector sample mean. Shrinkage toward a weighted means statistics appearing in random effects model. The results are illustrated by an example of collaborative studies.
Citation: IMS Lecture Notes-Mongraph Series
Pub Type: Others
Birge ratio, DerSimonian-Laird estimator, heteroscedasticity, key comparisons, meta-analysis, normal mean, reference value, shrinkage estimators