作者: Jian Dai
DOI: 10.1111/J.1538-4632.1998.TB00391.X
关键词:
摘要: Conventional discrete choice models assume implicitly that the set is independent of decisionmaker's preferences conditional on explanatory variables models. This assumption implausible in many situations where decisionmaker selects his or her set. paper estimates and tests a model with endogenous sets based Horowitz' theoretical work. To calibrate model, new probability simulator introduced sequential estimation procedure developed. The calibration methods are tested an empirical application as well Monte Carlo simulations. results used to test theory examine differences between conventional parameter predicted probabilities. strongly suggest ignoring endogeneity modeling can have serious consequences applications.