【主讲】马骏 (中国人民大学)
【主题】"Inference Using the Two-Step Nonparametric Estimator for First-Price Auction Models" (jointly with V. Marmer and A. Shneyerov)
【时间】2015年10月23日 (周五) 15：30-17：00
【地点】上海财经大学经济学院楼501室
【语言】英文
【摘要】In this paper, we focus on inference on the probability density function (PDF) of the private values in the first-price sealed-bid auction models within the independent private value paradigm. We show the asymptotic normality of the two-step nonparametric estimator of Guerre et al. (2000). We compare the asymptotic variance with that of the quantile-based one-step estimator of Marmer et al. (2012). By the comparison we achieve a result on the asymptotic relative efficiency of the methods and show that the two-step estimator is more efficient. We propose a convenient estimator of the asymptotic variance. Based on these results, we achieve a new approach to inference. We conduct Monte Carlo experiments to show the finite-sample performance of our approach.
