Robustness Criteria for Block Designs: Vulnerability in the Event of Observation Loss

When?: Wednesday 25 May 2011, 14:00 to 15:00
Where?: 22AA04
Open to:: Staff, Students
Speaker:: Helen Warren (nee Thornewell) (Surrey)

Abstract: This presentation will demonstrate some interesting examples, in order to summarise some of the key findings and contributions from my PhD. My PhD research investigates into the robustness of incomplete block designs in the event of observation loss. If observations are lost during the course of an experiment, the design properties are changed. In particular, a block design, which is optimal at the start of an experiment may result in a disconnected eventual design, in which case the usual null hypothesis cannot be tested, since not all elementary treatment contrasts are estimable.

In order to guard against poor eventual designs, a Vulnerability Measure is introduced to determine the likelihood of a design becoming disconnected as a result of random observation loss. This measure aids in design selection, comparison and construction. Comparisons are made between the vulnerability and optimality of designs. The vulnerability method is compared with other scenarios of observation loss: the loss of whole blocks and the loss of up to three random observations. The coincidence of these different robustness rankings and the relative strengths and weaknesses of the criteria are discussed. Results are applied to special design classes, including complement and repeated BIBDs. Non balanced designs are also considered, including Regular Graph Designs and Nearly Balanced Designs.

Robustness Criteria for Block Designs: Vulnerability in the Event of Observation Loss

Stay Connected

Contact Us