EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
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, 10:30-12:00Room: 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
- Behavioural OR
- Mathematical Programming
- Complexity and Approximation
Status: accepted
Back to the list of papers