BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Plone.org//NONSGML plone.app.event//EN
X-WR-TIMEZONE:Europe/Berlin
BEGIN:VEVENT
SUMMARY:03.06.2026 - Yucheng Yang (University of Zurich and SFI)
DTSTART;VALUE=DATE:20260603
DTEND;VALUE=DATE:20260604
DTSTAMP:20260903T231724Z
UID:cd82459318dd408fb16fb94f040eb3bb@www.econ.uni-bonn.de
CREATED:20260304T095412Z
DESCRIPTION:We present a new approach to formulating and solving heterogen
 eous agent mod-\nels with aggregate risk. We replace the cross-sectional d
 istribution with low-dimensional\nprices as state variables and let agents
  learn equilibrium price dynamics directly from sim-\nulated paths. To do 
 so\, we introduce a structural reinforcement learning (SRL) method which\n
 treats prices via simulation while exploiting agents’ structural knowled
 ge of their own in-\ndividual dynamics. Our SRL method yields a general an
 d highly efficient global solution\nmethod for heterogeneous agent models 
 that sidesteps the Master equation and handles\nmodels traditional methods
  struggle with\, like those with nontrivial market-clearing con-\nditions.
  We illustrate the approach in the Krusell-Smith model\, the Huggett model
  with\naggregate shocks\, and a HANK model with a forward-looking Phillips
  curve\, all of which\nwe solve globally within minutes.
LAST-MODIFIED:20260330T104702Z
URL:https://www.econ.uni-bonn.de/macro/en/seminars/mef-seminar-summer-26/y
 ucheng-yang-university-of-zurich-and-sfi
END:VEVENT
END:VCALENDAR