Finite Sample Inference for the Maximum Score Estimand

• 成果类型: Article; Early Access
• 署名作者: Rosen, Adam M.; Ura, Takuya
• 署名单位: Duke University; University of California System; University of California Davis
• 刊物名称: REVIEW OF ECONOMIC STUDIES
• ISSN/ISSBN: 0034-6527
• DOI: 10.1093/restud/rdaf001
• 发表日期: 2025
• 关键词: binary response models; partial identification; bootstrap; computation; algorithm; criterion

摘要

We provide a finite sample inference method for the structural parameters of Manski's semiparametric binary response model under a conditional median restriction. This is achieved by exploiting distributional properties of observable outcomes conditional on the observed sequence of exogenous variables. Moment inequalities conditional on the size n sequence of exogenous covariates are constructed, and the proposed test statistic is a monotone function of violations of the corresponding sample moment inequalities. The critical value used for inference is provided by the appropriate quantile of a known function of n independent Bernoulli random variables and does not require the use of a cube root asymptotic approximation employing a point estimator of the target parameter. Simulation studies demonstrate favourable finite sample performance of the test in comparison to several existing approaches. Empirical use is illustrated with an application to the classical setting of transportation choice.