Identification and Estimation of Entry Games under Symmetry of Unobservables

This paper provides a new point identification and estimation for two-player entry games with complete information under a symmetric condition on unobservables. Neither equilibrium selection nor distributional assumption is required. In addition, a weaker support condition is used in comparison to the existing literature. Following the identification, this paper proposes a new estimator, that, unlike existing estimators, is shown to have root-n consistency. This paper further constructs a test statistic to test the symmetry condition. Monte Carlo evidence demonstrates that the estimator performs well in moderate-sized samples and is robust to unimodal and multimodal error distributions. Finally, this paper applies the new method to the entry game of discount retailers in Jia (2008)