2002 World Congress of Environmental and Resource Economists

A recent nonmarket valuation survey used a question format called the iterative paired approach (Magat et al., 2000). Although it is the only instance of this format in the literature, the question format combines several features of other question formats commonly in use, namely the conjoint approach and the dichotomous-choice with follow-up questions. The combination of features, however, introduces some issues in econometric estimation, which are discussed here. A maximum likelihood model is proposed for this question format. Using a Monte Carlo simulation, the maximum likelihood model is then tested against a regression model that derives from a different interpretation of the questioning format. The maximum likelihood model is both unbiased and efficient, while the regression model is neither. This exercise also sheds light on a feature of the iterative paired approach that has received comparatively little attention in the stated preference literature: how an expression of indifference is to be interpreted.



	Paper Title:	Bias And Efficiency In The Iterative Paired Approach To Nonmarket Valuation
	Presenting Author:	Julie Hewitt (U.S. EPA)
	Coauthor 1:
	Coauthor 2:
	Coauthor 3:
	Abstract:	A recent nonmarket valuation survey used a question format called the iterative paired approach (Magat et al., 2000). Although it is the only instance of this format in the literature, the question format combines several features of other question formats commonly in use, namely the conjoint approach and the dichotomous-choice with follow-up questions. The combination of features, however, introduces some issues in econometric estimation, which are discussed here. A maximum likelihood model is proposed for this question format. Using a Monte Carlo simulation, the maximum likelihood model is then tested against a regression model that derives from a different interpretation of the questioning format. The maximum likelihood model is both unbiased and efficient, while the regression model is neither. This exercise also sheds light on a feature of the iterative paired approach that has received comparatively little attention in the stated preference literature: how an expression of indifference is to be interpreted.
	Link to paper:	Not available

	Session / Day / Time	15E / Thursday / 10:15 - 11:45 am

	Please use your browser's "back" button to get to the previous page.