Nonparametric two-stage estimation of conditional choice probabilities in a binary choice model under uncertainty

In this paper, estimation of conditional choice probabilities in a binary choice model under uncertainty is considered under weak nonparametric restrictions on expectations and the error distribution. The estimation method follows a two-stage strategy: the first stage estimates expectations using realizations of agents' future, and the second stage estimates conditional choice probabilities using choice data and the expectations estimates. This paper gives conditions under which the two-stage nonparametric kernel estimator is uniformly, strongly consistent and asymptotically normal.