This paper proposes a framework for deriving early-warning models with optimal out-of-sample forecasting properties and applies it to predicting distress in European banks.
The recent global financial crisis highlighted the large costs for societies that the unravelling of macro-financial imbalances can have. The corresponding policy response in the aftermath of the financial crisis has been to strengthen micro-prudential regulation of financial institutions and provide new macro-prudential mandates and tools to competent authorities with the aim of dampening financial cycles and making the financial system more resilient to adverse shocks.
A key issue for implementing macro-prudential policy is to identify the build-up of macro-financial vulnerabilities with a sufficient lead time so that policy action can still be effective in preventing severe financial crises. The policy interest in so-called early-warning models has therefore increased considerably in recent years, especially in the context of guiding the activation of macro-prudential policy tools.
At the same time the academic interest in early-warning models has also increased considerably recently, as various papers have shown that there indeed seem to be common patterns in the data that often precede financial crises.
Despite many previous efforts, building an early-warning model is a complex task that involves numerous assumptions and practical choices that need to be made.
The main contributions of the paper are threefold:
-
First, the paper introduces a conceptual framework to guide the process of building early-warning models, which highlights and structures the numerous complex choices that the modeler needs to make.
-
Second, the paper proposes a flexible modeling solution to the conceptual framework that supports model selection in real-time. Specifically, their proposed solution is to combine the loss function approach to evaluate early-warning models with regularized logistic regression and cross-validation to find a model specification with optimal real-time out-of-sample forecasting properties.
-
Third, the paper illustrates how the modeling framework can be used in analysis supporting both microand macro-prudential policy by applying it to a large dataset of EU banks and showing some examples of early-warning model visualizations.
Full working paper
© ECB - European Central Bank
Key
Hover over the blue highlighted
text to view the acronym meaning
Hover
over these icons for more information
Comments:
No Comments for this Article