BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//jEvents 2.0 for Joomla//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:2aa99d8fb4e6803f4a21d5b9006a2976
CATEGORIES:Seminars
SUMMARY:Lunch Seminar: Stéphane Bonhomme - University of Chicago
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:\n\nPosterior Average Effects (joint with Martin Weidner)\n\n\nAbstract:\nT
he applied economist is often interested in computing an average with respe
ct to a distribution of unobservables. Common examples are average partial
effects in discrete choice, moments of individual fixed-effects in panel da
ta, or counterfactual policy simulations based on a structural model. We co
nsider empirical Bayes (EB) estimators of such effects, where the average i
s computed conditional on the observation sample. While EB estimators are s
ometimes used to “shrink” individual estimates – e.g., of teacher value-add
ed or hospital quality – toward a common mean and reduce estimation noise,
a study of their frequentist properties is still lacking. We establish two
robustness properties of EB estimators under misspecification of the assume
d distribution of unobservables: EB estimators are optimal in terms of loca
l (worst-case) bias, and their global bias is no larger than twice the mini
mum bias that can be achieved within a large class of estimators of average
effects. These results provide a rationale for the use of empirical Bayes
estimators, beyond the settings to which they have been applied so far.\n
DTSTAMP:20200815T172117Z
DTSTART:20190305T123000Z
DTEND:20190305T133000Z
SEQUENCE:0
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR