Dr. Peter Sarlin

Dr. Peter Sarlin
Aktuelle Position

seit 2/17

Vorstandsvorsitzender und leitender Wissenschaftler

SILO.AI

seit 1/15

Research Affiliate

Leibniz-Institut für Wirtschaftsforschung Halle (IWH)

seit 1/14

Professor

Hanken School of Economics

Leiter

RiskLab Finland

Forschungsschwerpunkte

  • Analyse systemischen Risikos
  • makroprudenzielle Politik
  • maschinelles Lernen und visuelle Analyse

Peter Sarlin ist seit Januar 2015 Research Affiliate am IWH. Er forscht zu den Themen maschinelles Lernen und künstliche Intelligenz.

Peter Sarlin ist Professor an der Hanken School of Economics, Direktor von SILO.AI und Leiter von RiskLab Finland.

Ihr Kontakt

Dr. Peter Sarlin
Dr. Peter Sarlin
Mitglied - Abteilung Makroökonomik
Nachricht senden Persönliche Seite

Publikationen

cover_macroeconomic-dynamics.jpg

Optimizing Policymakers’ Loss Functions in Crisis Prediction: Before, Within or After?

Peter Sarlin Gregor von Schweinitz

in: Macroeconomic Dynamics, im Erscheinen

Abstract

Early-warning models most commonly optimize signaling thresholds on crisis probabilities. The expost 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-à-vis missing a crisis. We provide simulated and real-world evidence that this simplification results in stable thresholds and improves out-of-sample performance. Our solution is not restricted to binary-choice models, but directly transferable to the signaling approach and all probabilistic early-warning models.

Publikation lesen
Mitglied der Leibniz-Gemeinschaft LogoTotal-Equality-LogoWeltoffen Logo