The purpose of this paper is to study the estimation in some experimental plans where the choice of experimental units depends on what is observed in the before period. In section 2, some notations are introduced and a model for experiments studied is described. In the third section examples of 3 special models that will be treated in more detail, are given. Section 4 discusses estimation procedures. In the 5th section, the estimation procedures are applied to 3 special cases. Asymptomic properties of the estimates are derived. Some simulations are presented in the last section.
Abstract