Michael Baron, PhD
Professor, Department of Mathematical Sciences
University of Texas at Dallas, Richardson, Texas
|
Optimal Sequential Planning**
Abstract:
In clinical trials, it is often impractical and expensive to sample according to the classical sequential scheme, that is, one observation at a time. Group sequential methods allow to take batches of observations. A rather recent methodology of sequential planning extends and generalizes the group sequential methods by choosing the sample sizes sequentially. We discuss optimality of sequential designs taking into account both the variable cost and the fixed cost of experiments. General guidelines for optimal sequential planning are established. Direct computations methods for the risk evaluation are developed for sequential planning on a lattice.
|
Location: |
AE 1002 (Biostatistics Seminar Room - Pavilion I) |
|
Date: |
Thursday, July 30, 2009 |
|
Time: |
3:00 – 4:00 PM |
|
Contact: |
Lifang Zhang
(706) 721-4453 or Biostat@MCG.edu |
|
