The more uncertain cost-effectiveness estimates are, the stronger the effect of optimizer’s curse is. Hence we should prefer interventions whose cost-effectiveness estimates are more robust.
Faced with such uncertainty, shouldn’t we rather hedge our bets and split our support? For example, if project A has cost-effectiveness in the range of 70-80%, and project B has cost-effectiveness in the range 60-90%, wouldn’t it be better (overall) to split the support evenly than to only support project A?
Faced with such uncertainty, shouldn’t we rather hedge our bets and split our support? For example, if project A has cost-effectiveness in the range of 70-80%, and project B has cost-effectiveness in the range 60-90%, wouldn’t it be better (overall) to split the support evenly than to only support project A?