Abstract
We consider the situation where a number of treatments are successively applied to a single experimental unit with both the direct effect of a treatment (in the period it is applied) and its residual effect (in the following period) being considered as two separate components of variation. We specialize to the case where the residual effects are certain known/unknown fractions of the correeponding direct effects. We discuss a method of analysis of the underlying data and also present various related results.
Get full access to this article
View all access options for this article.