1: \begin{abstract}

2: \thispagestyle{empty}

3: 

4: We propose a semiparametric

5: method to estimate the density of private values in first-price auctions. 

6: Specifically, we

7: model private values through a set of conditional

8: moment restrictions and use a two-step procedure.

9: In the first step we recover a sample of pseudo

10: private values using Local Polynomial Estimator. In the second step

11: we use a GMM procedure to estimate the parameter(s) of interest. We show that the proposed semiparametric estimator is consistent, has an asymptotic normal distribution, and attains the parametric (``root-n'') rate of convergence. 

12: 

13: 

14: \bigskip

15: 

16: \noindent Keywords: Empirical Auctions, Semiparametric Estimator, Local Polynomial, GMM.\\

17: JEL Codes: C14, C71, D44.

18: 

19: \end{abstract}

20: