Efficient Circular Cross-over Designs for Models with Interaction
Cross-over designs are widely used in many areas such as clinical trials or agricultural field experiments. In addition to the effect of the treatment applied to a given period, a residual effect of the previous treatment is often observed. In that case, the effect observed depends on both treatments. When the aim of the experiment is to select a single treatment, it is natural to consider the effect of a treatment preceded by itself, also called the total effect. We consider universally optimal or efficient circular cross-over designs for total effects under a model with interaction between treatment and carry-over effects. The optimal designs are obtained from the theory of approximate designs developed by Kushner (Ann Stat 25:2328–2344, 1997).
- 12.Kiefer, J.: Construction and optimality of generalized Youden designs. In: A Survey of Statistical Design and Linear Models. Proceedings of International Symposium, Colorado State University, Ft. Collins, 1973, pp. 333–353. North-Holland, Amsterdam (1975)Google Scholar