BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:iCalendar-Ruby
BEGIN:VEVENT
CATEGORIES:Class/ Workshop
DESCRIPTION:Stéphane Bonhomme - University of Chicago - CANCELLED\n\nPoster
ior Average Effects (joint w/Martin Weidner)\n\nAbstract: Economists are of
ten interested in estimating averages with respect to distributions of unob
servables. Examples are moments of individual fixed-effects\, average effec
ts in discrete choice models\, or counterfactual simulations in structural
models. For such quantities\, we propose and study “posterior average effec
ts”\, where the average is computed conditional on the sample\, in the spir
it of empirical Bayes and shrinkage methods. While the usefulness of shrink
age for prediction is well-understood\, a justification of posterior condit
ioning to estimate population averages is currently lacking. We establish
two robustness properties of posterior average effects under misspecificati
on of the assumed distribution of unobservables: they are optimal in terms
of local worst-case bias\, and their global bias is at most twice the minim
um worst-case bias within a large class of estimators. We establish related
robustness results for posterior predictors. In addition\, we suggest a si
mple measure of the information contained in the posterior conditioning. La
stly\, we present two empirical illustrations\, to estimate the distributio
ns of neighborhood effects in the US\, and of permanent and transitory comp
onents in a model of income dynamics.
DTEND:20200317T171000Z
DTSTAMP:20210920T114903Z
DTSTART:20200317T154000Z
GEO:42.447296;-76.482254
LOCATION:Uris Hall\, 498
SEQUENCE:0
SUMMARY:CANCELLED - Econometrics Workshop: Stéphane Bonhomme
UID:tag:localist.com\,2008:EventInstance_32194372381542
URL:https://events.cornell.edu/event/econometrics_workshop_stephane_bonhomm
e_-_university_of_chicago
END:VEVENT
END:VCALENDAR