European Central Bank - eurosystem
Opțiuni de căutare
Pagina inițială Media Materiale explicative Studii și publicații Statistici Politică monetară Euro Plăți și piețe Cariere
Sortează în funcție de
Nu este disponibil în limba română

Gregor von Schweinitz

23 February 2017
Early-warning models most commonly optimize signaling thresholds on crisis probabilities. The ex-post threshold optimization is based upon a loss function accounting for preferences between forecast errors, but comes with two crucial drawbacks: unstable thresholds in recursive estimations and an in-sample overfit at the expense of out-of-sample performance. We propose two alternatives for threshold setting: (i) including preferences in the estimation itself and (ii) setting thresholds ex-ante according to preferences only. Given probabilistic model output, it is intuitive that a decision rule is independent of the data or model specification, as thresholds on probabilities represent a willingness to issue a false alarm vis-
JEL Code
C35 : Mathematical and Quantitative Methods→Multiple or Simultaneous Equation Models, Multiple Variables→Discrete Regression and Qualitative Choice Models, Discrete Regressors, Proportions
C53 : Mathematical and Quantitative Methods→Econometric Modeling→Forecasting and Prediction Methods, Simulation Methods
G01 : Financial Economics→General→Financial Crises