ABSTRACT
In this chapter, we consider numerous further applications of the MOVER and PropImp algorithms. These include a method for calculating confidence intervals for interaction effects on an additive scale that has been comprehensively evaluated. Often, the quantity of interest f (X, Y) can be expressed in the form g( f1(X) – f2(Y)) or g( f1(X)/f2(Y)) for some functions f1, f2 and g. In such situations the MOVER approach is easy to apply. This is the case for many but not all the examples considered here. In several situations this cannot be done, and the PropImp algorithm is needed.
