Kevin Dano - Princeton University

Abstract

Many panel data methods, while allowing for general dependence between covariates and time-invariant agent-specific heterogeneity, place strong a priori restrictions on feedback: how past outcomes, covariates, and heterogeneity map into future covariate levels. Ruling out feedback, as is typically done, is restrictive in many dynamic eco- nomic settings. We provide a general characterization of all “feedback robust” moment conditions for nonlinear panel data models and present constructive methods to derive feasible estimators in specific models. We also use our moment characterization to compute semi-parametric efficiency bounds, allowing for a quantification of the information loss associated with accommodating feedback. We present applications of our methods to linear and nonlinear regression models as well as to duration models. For the latter case, we present new results which relax the proportional hazards assumption.

Additional information:

• Speaker: Kevin Dano
• Time: Thursday, 14.11.2024, 16:00 - 17:00
• Location: Faculty Meetingroom, U 1.040
• Further links:
  • Kevin Dano's webpage
• Organizer: Statistics Group
• Contact:
  • Almut Lunkenheimer
  • +49 228 73-9228
  • ifs@uni-bonn.de