Fat tails and truncated bids in contingent valuation: an application to an endangered shorebird species
Fat tails in contingent valuation (CV) refers to the phenomenon of a yesresponse function having a high and slowly declining yes-response rate at high bid levels offered in a CV survey. So, for example, a yes-response rate might hold at 20% or greater over the three or four highest bids offered in a...
- Autores:
- Tipo de recurso:
- Part of book
- Fecha de publicación:
- 2017
- Institución:
- Universidad de Bogotá Jorge Tadeo Lozano
- Repositorio:
- Expeditio: repositorio UTadeo
- Idioma:
- eng
- OAI Identifier:
- oai:expeditiorepositorio.utadeo.edu.co:20.500.12010/15499
- Acceso en línea:
- http://hdl.handle.net/20.500.12010/15499
- Palabra clave:
- Endangered shorebird species
Aves raras
Especies en vías de extinción
Aves
- Rights
- License
- Abierto (Texto Completo)
| Summary: | Fat tails in contingent valuation (CV) refers to the phenomenon of a yesresponse function having a high and slowly declining yes-response rate at high bid levels offered in a CV survey. So, for example, a yes-response rate might hold at 20% or greater over the three or four highest bids offered in a survey. The “tails” of the yes-response function are said to be “fat” in this case. A truncated bid refers to a circumstance where high bids are not offered over a range where it appears as though the survey instrument would produce a non-zero percentage of yes responses – essentially ignoring the behavioral response to high bids or “truncating” the yes-response function. Fat tails has been recognized and discussed in the CV literature for more than two decades (Desvousges et al., 1993). Analysts have also recognized that fat tails can create problems for parametric estimators (e.g., logit and probit), wherein the estimators are sensitive to the highest bids offered in a survey (Cooper and Loomis, 1992; Desvousges et al., 1993). In part because of this problem and in part because of the problem of negative willingness-to-pay (WTP) estimates from parametric estimators, the field has turned toward non-parametric estimators, especially the Turnbull lower bound (Kriström, 1990; Carson et al., 1994; Haab and McConnell, 1997). This chapter shows that fat tails also create problems for non-parametric estimators. The real issues presented in the data do not go away by simply changing estimators. |
|---|