SMU SOE Seminar (Mar 10, 2017): Adaptive Inference in Semiparametric Multinomial Response Models

We consider identification, estimation and inference on regression coeffcients in semi parametric multinomial response models. Our identification result is constructive and estimation is based on a localized rank objective function, loosely analogous to that used in Abrevaya, Hausman, and Khan (2010). We we show this achieves sharp identification which is in contrast to existing procedures in the literature such as, for example, Ahn, Powell, Ichimura, and Ruud (2014). In that sense, our procedure is adaptive (Khan and Tamer (2009)) in the sense that it provides an estimator of the sharp set when point identification does not hold, and a consistent point estimator when it does. Furthermore, our rank procedure extends to panel data settings for inference in models with fixed effects, including dynamic panel models with lagged dependent variables as covariates. A simulation study establishes adequate finite sample properties of our new procedures.

Click here to view his CV.