Distribution theory for the analysis of binary choice under uncertainty with nonparametric estimation of expectations (Q1209890): Difference between revisions

The paper deals with the estimator of preferences previously introduced by Manski, analyzing its asymptotic sampling behavior. \textit{C. F. Manski} [Nonparametric and semiparametric methods in econometrics and statistics. Proc. 5th Int. Symp., Econ. Theory Econ., Durham/NC (USA) 1988, 259-275 (1991; Zbl 0757.62018)] introduced a two-stage, semiparametric estimator applicable if expectations are fulfilled and are conditioned only on variables observed by the researcher. Manski's estimator leads to a consistent estimator of preference parameters if the population is sufficiently regular and if the expectations estimate is chosen appropriately. The present paper solves a further question referred to the influence that a nonparametric estimation of expectations has in the parametric estimation of preferences. After setting out the estimation problem and presenting the estimator, a theorem is proved showing that, given six assumptions, the preference estimate enjoys the classical asymptotic properties of \(\sqrt{N}\)- consistency, \(\sqrt{N}\)-asymptotic unbiasedness and asymptotic normality. According to my opinion this paper is an interesting new contribution to the asymptotic analysis of preferences estimation.