ABSTRACT
People differ widely in how they react to events. Most scientific studies express the effect of a treatment as an average over a group of persons. This is informative if the effect is thought to be similar for all persons, but is less useful if the effect is expected to differ. This chapter uses multiple imputation to estimate the individual causal effect (ICE), or the unit-level causal effect, for one or more units in the data. The hope is that this allows us to develop a deeper understanding of how and why people differ in their reactions to an intervention.
