Killeen Peter R
Department of Psychology, ARizona STate University, Tempe, AZ 85287-1104, USA.
Psychol Rev. 2009 Jul;116(3):602-19. doi: 10.1037/a0016414.
Goods remote in temporal, spatial, or social distance, or in likelihood, exert less control over our behavior than those more proximate. The decay of influence with distance, of perennial interest to behavioral economists, has had a renaissance in the study of delay discounting. By developing discount functions from marginal utilities, this article provides a framework that resolves several anomalies of intertemporal choice. Utilities are inferred to be power functions of monetary value, delay, and probability. Utility, not value, is discounted, with decisions made by adding the utility of a good to the disutility of a delay or contingency. The theory reduces to standard treatments, such as exponential, hyperbolic and hyperboloid, and exponential-power; naturally predicts magnitude effects and other asymmetries; is consistent with subadditivity, immediacy, and certainty effects; returns conjointly measured determinations of monetary utility and temporal distance functions; and is extensible to other dimensions of distance.
在时间、空间或社会距离上较为遥远,或在可能性上较低的事物,对我们行为的控制要比那些更接近的事物弱。距离导致的影响力衰减一直是行为经济学家长期感兴趣的问题,在延迟折扣研究中再度兴起。通过从边际效用发展出折扣函数,本文提供了一个解决跨期选择中若干异常现象的框架。效用被推断为货币价值、延迟和概率的幂函数。被贴现的是效用而非价值,决策是通过将一个物品的效用与延迟或意外情况的负效用相加来做出的。该理论简化为标准处理方式,如指数、双曲线和双曲面以及指数幂;自然地预测了量级效应和其他不对称性;与次可加性、即时性和确定性效应一致;得出货币效用和时间距离函数的联合测量结果;并且可扩展到距离的其他维度。