Two-stage sampling in pharmaceutical applications.

In the pharmaceutical industry, some tests for quality assurance before, during and after the manufacture of a drug product involve a two-stage sampling technique. We propose statistical testing procedures in a two-stage sampling problem with large within-class sample sizes. We also derive a two-stage sampling plan by minimizing the expected squared volume (or the generalized variance) of the confidence region related to the test. We present two examples to illustrate the proposed methods.

Duke Scholars

Author Shein-Chung Chow Biostatistics & Bioinformatics, Division of Biostatistics