Obtaining Welfare Bounds in Discrete-Response Valuation Studies: Comment
评论Boman等人1999年提出的从离散响应条件价值评估数据中提取平均福利变化非参数界限的方法，指出其虽能简化计算并处理不确定性，但需准确估计支付意愿均值的方差才能构建置信区间。
In a recent article published in Land Economics, Boman, Bostedt, and Kristrim (1999, hereafter BBK) show how nonparametric lower bound, intermediate and upper bound measures of average welfare change can be extracted from discrete-response contingent valuation survey data, building on the work of Kristrim (1990) and Haab and McConnell (1997). The article appears to offer a promising route to getting estimates of mean willingness to pay (WTP) along with their respective variances without having to estimate parametric choice models, thereby providing a computationally simple way to capture both the methodological uncertainty about the choice of a central tendency measure and the statistical uncertainty associated with any particular measure. Having an accurate estimate of the variance of mean WTP allows one to make confidence interval statements about mean WTP. When measures of mean WTP and its variance are extracted