Targeted Crisis Risk Control : A Neyman-Pearson Approach

This paper investigates the link between crisis risk estimation and crisis prevention policy in the context of sudden stops with growth impacts. We propose a two-stage framework that embeds an early-warning problem into a policy-making problem, and conduct a welfare analysis of crisis risk estimation based on subsequent policy responses. Building upon this two-stage framework, we show that there is welfare cost asymmetry between two types of errors: the welfare-maximizing weight on the missed crises is greater than that on the false alarms. This paper introduces a constrained optimization problem under the Neyman-Pearson paradigm to solve this error-asymmetry problem, and conducts backtesting with multiple algorithms under this paradigm with asymmetry-warranted evaluation metric. Bringing this welfare-based two-stage framework to emerging market countries' data, we uncover the time-varying risk tolerance of policymakers