The Estimation of Indifference Maps by Expected Utility Analysis
检验了Sinden用于估值非市场商品的无差异曲线估计方法，发现其严格可加性假设会导致偏差，并提出一种非可加替代方法，但两种方法均受限于受访者不一致性，质疑了该方法的实用性。
Abstract The indifference curve estimation procedure proposed and used by Sinden to value nonmarketed goods implicitly adopts the assumption that a two‐good utility function can be obtained by the simple addition of two single‐good utility functions. The results of two experiments indicate that using the strict additivity assumption is likely to bias the indifference maps so derived. While an alternative nonadditive technique is proposed, it is found that, as with Sinden's technique, it is subject to such a large degree of respondent inconsistency that doubt is cast on the usefulness of the results obtained and on the practicality of this general approach.