Endogenous entry in first-price auctions

This thesis studies the first-price auction models with endogenous entry. In the first chapter, we propose using the nearest neighbor estimation technique to estimate the entry cost in the auction models. We study the large sample properties of the proposed estimator to establish sets of conditions sufficient for its consistency and asymptotic normality. To give an example, we further apply the proposed method to estimate the cost of participation in the Michigan Highway Procurement Auctions. Our study rejects the null hypothesis of zero participation costs. Based on our estimation result, we infer how the optimal auction outcomes can be realized by using the regular policy tools. We demonstrate the improvement that the Michigan government could have made on payments if the optimal auctions had been employed. In the second chapter, we propose applying the simulated non-linear least squares (SNLLS) method of Laffont, Ossard and Vuong (1995, LOV hereafter) [28] to estimate the distribution of bidders' private values in the estimation of a first-price auction model with endogenous entry. Unlike the estimation problem of LOV , however, the valuation distribution of bidders is truncated in our problem, due to the presence of the entry cost. This further causes the absence of continuous differentiability in our statistical estimation objective function, which was required in LOV's large sample analysis. Therefore, we provide a separate analysis to study the large sample behavior of the SNLLS estimator in such setup. In the third chapter, we develop a nonparametric method that allows one to discriminate among alternative models of entry in first-price auctions. Three models of entry are considered: Levin and Smith (1994) [31], Samuelson (1985) [53], and a new model in which the information received at the entry stage is imperfectly correlated with valuations. We show that these models impose different restrictions on the quantiles of active bidders' valuations, and develop nonparametric tests of these restrictions. We perform the tests using a dataset of highway procurement auctions in Oklahoma. Depending on the project size, we find somewhat more support for the new model.Arts, Faculty ofVancouver School of EconomicsGraduat