Choosing treatment policies under ambiguity

Economists studying choice with partial knowledge typically assume that the decision maker places a subjective distribution on unknown quantities and maximizes expected utility. Someone lacking a subjective distribution faces a problem of choice under ambiguity. This article reviews recent research on policy choice under ambiguity, when the task is to choose treatments for a population. Ambiguity arises when a planner has partial knowledge of treatment response and does not feel able to place a subjective distribution on the unknowns. I first discuss dominance and alternative criteria for choice among undominated policies. I then illustrate with the choice of a vaccination policy by a planner who has partial knowledge of the effect of vaccination on illness. I next study a class of problems in which a planner may want to cope with ambiguity by diversification, assigning observationally identical persons to different treatments. Lastly, I consider a setting in which a planner should not diversify treatment.