Stated or inferred attribute non-attendance? A simulation approach
DOI:
https://doi.org/10.7201/earn.2013.01.03Keywords:
Attribute non-attendance, choice modelling, environmental valuation, random parameter logit.Abstract
In the last few years, there has been a growing body of literature on how to detect and deal with the fact that some respondents seem to ignore one or more attributes in a discrete choice experiments.This paper aims to analyse the performance of two econometric approaches devoted to solve this problem: the stated attribute non-attendance approach and the inferred attribute non-attendance approach. These approaches are examined further by two common ways of collecting information on attribute non-attendance: serial and choice task non-attendance. The results of the simulation experiments show firstly, that choice task non-attendance of one attribute causes biases in the estimation of all other parameters; and, secondly, that only serial non-attendance can be inferred successfully. The results are policy relevant because not treating, or treating this issue incorrectly may end up in biased welfare measures.
Downloads
Published
2013-06-07
Issue
Section
Articles
License
Assignment of publishing rights for accepted manuscripts: Authors will assign to the Spanish Association of Agricultural Economics (Asociación Española de Economía Agraria, AEEA) the copyright of the manuscript and any supplemental tables, illustrations or other information submitted therewith that are intended for publication as part of or as a supplement to the manuscript in all forms and media (whether now known or hereafter developed), throughout the world, in all languages, for the full term of copyright, effective when and if the article is accepted for publication. No revisions, additional terms or addenda to this Agreement can be accepted without our express written consent.