Bias in Estimating the Almost Ideal Demand System with the Stone Index Approximation
证明常用的Stone指数近似会导致几乎理想需求系统的参数估计产生偏差，主要影响价格参数，在使用家庭微观数据时偏差更严重；若无强替代效应，可通过简单重新参数化纠正。
This paper shows that the commonly used Stone index approximation can bias the parameter estimates of the almost ideal demand system. The bias mostly affects the price parameters and can be more serious when the budget share equations are estimated with individual household rather than aggregate data. In the absence of strong substitution effects, the price parameter bias can be largely corrected through simple reparameterization. Copyright 1993 by Royal Economic Society.