Optimal Designs for Simultaneous Measurements in Compartment ModelsView Presentation *Jiewei Zeng,
Keywords: compartment models, General Equivalence Theorem, Tchebycheff system
In pharmaceutical industries, modeling of dynamic systems plays a very important role in applications. Among many others, the estimates of unknown parameters (related to ADME, toxicity and efficacy) in these models are highly important in drug developments. Good estimates rely on the experimental design, that is, the optimal number of design points, and the number of samples to be collected at each design point. The compartment models, which are among the most useful tools for analyzing dynamic systems, have been mainly studied here. To achieve the accuracy, people usually collect several measurements at one design point. Different from the traditional i.i.d sampling, we introduce correlation structure to these measurements. The well-known General Equivalence Theorem by Kiefer and Wolfowitz has been adapted to this case and Tchebycheff system has been used to find the optimal number of support points for these models. Numerical search for the design points has also been conducted for practical use.