Title: Accounting for attribute non-attendance in the analysis of discrete choice experiments
Authors: Meulders, Michel
Rousseau, Sandra
Vermunt, Jeroen
Vranken, Liesbet
Issue Date: Dec-2014
Conference: International Conference of the ERCIM Working Group on Computational and Methodological Statistics edition:7 location:Pisa date:6-8 December 2014
Abstract: Nowadays stated preference techniques such as discrete choice experiments have become a popular tool to model the product preference of consumers as a function of product characteristics. Data from discrete choice experiments are often modeled with latent class conditional logit models to account for the fact that consumers may weigh product characteristics differently when choosing among products. Furthermore, a standard assumption when analyzing data from choice experiments is that subjects consider all attributes as relevant when choosing the most preferred alternative from a choice set. However, research has indicated that consumers may only attend to specific subsets of attributes when choosing between alternatives. This so-called attribute nonattendance can emerge for several reasons and failure to take it into account may lead to biased marginal willingness-to-pay estimates for product characteristics. In this paper we discuss how standard latent class conditional logit models (i.e. with class-specific regression parameters) can be extended to account for attribute non-attendance. As an illustration, the models are applied to analyze consumers’ willingness-to-pay for the organic label on apples.
Publication status: published
KU Leuven publication type: IMa
Appears in Collections:Research Centre for Quantitative Business Processes, Campus Brussels (-)
Faculty of Economics and Business (FEB) - miscellaneous
Research Centre for Economics and Corporate Sustainability, Campus Brussels
Faculty of Social Sciences - miscellaneous
Division of Bioeconomics

Files in This Item:

There are no files associated with this item.


All items in Lirias are protected by copyright, with all rights reserved.