Fast and Simple Adaptive Elicitations

53 Pages Posted: 5 May 2020 Last revised: 24 Feb 2022

See all articles by Nicolò Bertani

Nicolò Bertani

Catholic University of Portugal (UCP) - Catolica Lisbon School of Business and Economics

Abdellah Boukhatem

affiliation not provided to SSRN

Enrico Diecidue

INSEAD – Decision Sciences

Patrice Perny

affiliation not provided to SSRN

Paolo Viappiani

affiliation not provided to SSRN

Date Written: June 9, 2021

Abstract

We propose a new adaptive procedure for the measurement of decision models. Our procedure bounds the function of the decision model, sequentially halves the maximal distance between its bounds, and incrementally restricts the feasible space of associated parameters. It relies on approximating splines that improve tractability and descriptiveness. We apply and illustrate our procedure to elicit the probability weighting function in a simulation study and laboratory experiment. Our procedure faithfully recovers the underlying function and captures empirical regularities in probability weighting. In addition, it reveals systematic distortions introduced by standard parametric forms and sheds new light on the pervasiveness of possibility and certainty effects.

Keywords: elicitations, Expected Utility, Rank-Dependent Utility, probability weighting function

JEL Classification: C1

Suggested Citation

Bertani, Nicolò and Boukhatem, Abdellah and Diecidue, Enrico and Perny, Patrice and Viappiani, Paolo, Fast and Simple Adaptive Elicitations (June 9, 2021). Available at SSRN: https://ssrn.com/abstract=3569625 or http://dx.doi.org/10.2139/ssrn.3569625

Nicolò Bertani

Catholic University of Portugal (UCP) - Catolica Lisbon School of Business and Economics ( email )

Palma de Cima
Lisbon, 1649-023
Portugal

Abdellah Boukhatem

affiliation not provided to SSRN

Enrico Diecidue (Contact Author)

INSEAD – Decision Sciences ( email )

France

Patrice Perny

affiliation not provided to SSRN

Paolo Viappiani

affiliation not provided to SSRN

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
101
Abstract Views
558
rank
360,344
PlumX Metrics