Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/154909
Authors: 
Scarpa, Riccardo
Bateman, Ian J.
Year of Publication: 
1998
Series/Report no.: 
Nota di Lavoro, Fondazione Eni Enrico Mattei 51.1998
Abstract: 
Positive response density estimation from CV interval data affords efficiency gains which must be weighed against the risk of introducing potential bias during questions iteration. This study examines the effect of eliciting a third response on a set of often-used welfare measures derived in a conventional parametric setting. It then compares these with distribution-free nonparametric estimates. A third bound increases censoring probability, introduces welfare estimates sensitivity to inclusion of a theoretically relevant covariate such as wealth which also affects efficiency gains. This might well introduce complications that outweigh the expected efficiency gain. This empirical finding supports and complements previous results obtained via simulation.
Subjects: 
Parametric
Nonparametric Welfare measures
Contingent valuation (CV)
Willingness to pay (WTP)
JEL: 
C14
C42
H41
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.