EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

377. Forcasting behaviour in instable environments

Invited abstract in session WB-40: Experimental economics and game theory 1, stream Experimental economics and game theory.

Wednesday, 10:30-12:00
Room: 96 (building: 306)

Authors (first author is the speaker)

1. Ulrike Leopold-Wildburger
Department of Operations and Information Systems, University of Graz

Abstract

Time series models face structural instability when applied to real data. Usually, studies yield to applications without the consideration of breaks leads to unreliable results.
Progress has been achieved in the theory of identifying, estimation and testing of structural instabilities. Our experimental study shows who efficient the so-called bounds & likelihood heuristic can be. The subjects quickly identify breaks and are able to adapt their forecasts.

Time series models face structural instability when applied to real data. Usually, studies yield to applications without the consideration of breaks leads to unreliable results.
Progress has been achieved in the theory of identifying, estimation and testing of structural instabilities by our new procedure. The experimental studies show how efficient the so-called bounds & likelihood heuristic can be. The subjects quickly identify breaks and are able to adapt their forecasts in a surprisingly good manner. The subjects can be well explained by the b&l model despite the occurrence of structural breaks.
The bounds & likelihood heuristic manages to model average (not individual) forecasts.

Keywords

Status: accepted


Back to the list of papers