Parametric Recoverability of Preferences

Abstract

Revealed preference theory is brought to bear on the problem of recovering approximate parametric preferences from consistent and inconsistent consumer choices. We propose measures of the incompatibility between the revealed preference ranking implied by choices and the ranking induced by the considered parametric preferences. These incompatibility measures are proven to characterize well-known inconsistency indices. We advocate a recovery approach that is based on such incompatibility measures, and demonstrate its applicability for misspecication measurement and model selection. Using an innovative experimental design we empirically substantiate that the proposed revealed-preference-based method predicts choices signicantly better than a standard distance-based method. * We are grateful to the editor Ali Hortaçsu and anonymous referees for insightful comments and suggestions that signicantly improved this work. We thank Jose Apesteguia,

Homothetic Efficiency: Theory and Applications

“…This is motivated by a concept called e-and h-rationalization, which was recently introduced by Halevy, Persitz, and Zrill (2015): for efficiency close to 100%, there still exists a utility function that adequately explains the data as the result of homothetic utility maximization with only minor deviation.…”

“…Theorem 3 follows from Theorem 1 in Halevy, Persitz, and Zrill (2015). As their theorem does not contain our condition 2 (it can be constructed from the details of their proof), we include a brief proof of our Theorem 3 in the Appendix.…”

“…The equivalence of (1) and (3) is part of Theorem 1 by Halevy, Persitz, and Zrill (2015). The inequalities of condition (2) can be constructed from the details of their proof.…”

“…Therefore, to have a more detailed and robust data analysis, we extend this measure by introducing the more disaggregated homothetic efficiency vector (HEV), which provides efficiency indices for each observed choice. We provide a strong theoretical justification for using the HEI and HEV based on the concept of vector-rationalization recently introduced by Halevy, Persitz, and Zrill (2015).…”

Homothetic Preferences Revealed

“…Our method allows for a more detailed graphical analysis to reveal subjects' preferences and to choose appropriate functional forms for parametric analysis. It can also be used to extend applications of the money metric function, such as parametric recoverability as introduced by Halevy et al (2017). It can also improve nonparametric comparison of preferences as suggested by Heufer (2014).…”