Statistical decision functions with judgment

Abstract

A decision maker tests whether the gradient of the loss function evaluated at a judgmental decision is zero, for a given level of significance. If the test does not reject, the decision maker selects the judgmental decision. If the test rejects, the decision maker chooses the action whose gradient is at the boundary of the rejection region. The test is admissible and asymptotically most powerful. The level of significance reflects the decision maker's attitude toward uncertainty. The decision rule is applied to a problem of asset allocation.