BEGIN:VCALENDAR VERSION:2.0 PRODID:-//jEvents 2.0 for Joomla//EN CALSCALE:GREGORIAN METHOD:PUBLISH BEGIN:VEVENT UID:2aa99d8fb4e6803f4a21d5b9006a2976 CATEGORIES:Seminars CREATED:20190228T101633 SUMMARY:Lunch Seminar: Stéphane Bonhomme - University of Chicago DESCRIPTION;ENCODING=QUOTED-PRINTABLE:
Posterior Average Effects (joint with Martin Weidner) p>
Abstract:
The applied economist is often interested in computing an average with respect to a distribution of unobservables. Common examples are average partial effects in discrete choi ce, moments of individual fixed-effects in panel data, or counterfactual po licy simulations based on a structural model. We consider empirical Bayes ( EB) estimators of such effects, where the average is computed conditional o n the observation sample. While EB estimators are sometimes used to “shrink ” individual estimates – e.g., of teacher value-added or hospital quality – toward a common mean and reduce estimation noise, a study of their frequen tist properties is still lacking. We establish two robustness properties of EB estimators under misspecification of the assumed distribution of unobse rvables: EB estimators are optimal in terms of local (worst-case) bias, and their global bias is no larger than twice the minimum bias that can be ach ieved within a large class of estimators of average effects. These results provide a rationale for the use of empirical Bayes estimators, beyond the s ettings to which they have been applied so far.
DTSTAMP:20240329T145424Z DTSTART:20190305T133000Z DTEND:20190305T143000Z SEQUENCE:0 TRANSP:OPAQUE END:VEVENT END:VCALENDAR