RCBDs with a control |
| |
Authors: | Xiaowei Wang JP Morgan |
| |
Institution: | a Statistical Sciences (North America), Glaxo Smith Kline, King of Prussia, PA 19406, USA b Department of Statistics, Virginia Tech, Blacksburg, VA 24061-0439, USA |
| |
Abstract: | The randomized complete block designs, RCBDs, are among the most popular of block designs for comparing a set of experimental treatments. The question of this design's effectiveness when one of the treatments is a control is examined here. Optimality ranges are established for the RBCD in terms of the strength of interest in control comparisons. It is found that if the control treatment is of secondary interest, the RCBD, when not best, is typically near best. This is not so when comparisons with the control are of greater interest than those among the other treatments. |
| |
Keywords: | Blocking Control treatment Efficient design Optimal design Weighted optimality |
本文献已被 ScienceDirect 等数据库收录! |
|