BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250910T060910EDT-77163NdFIk@132.216.98.100
DTSTAMP:20250910T100910Z
DESCRIPTION:Mathieu Marcoux (Université de Montréal)\n	Host: Laura Lasio\n\n
 Date: March 13\, 2020\n	Location: Leacock 429\n\n“Imposing equilibrium rest
 rictions in the estimation of dynamic discrete games'\n\nAbstract:\n	Imposi
 ng equilibrium restrictions provides substantial gains in the estimation o
 f dynamic discrete games. Estimation algorithms imposing these restriction
 s – MPEC\, NFXP\, NPL\, and variations – have different merits and limitat
 ions. MPEC guarantees local convergence\, but requires the computation of 
 high-dimensional Jacobians. The NPL algorithm avoids the computation of th
 ese matrices\, but – in games – may fail to converge to the consistent NPL
  estimator. We study the asymptotic properties of the NPL algorithm treati
 ng the iterative procedure as performed in finite samples. We find that th
 ere are always samples for which the algorithm fails to converge\, and thi
 s introduces a selection bias. We also propose a spectral algorithm to com
 pute the NPL estimator. This algorithm satisfies local convergence and avo
 ids the computation of Jacobian matrices. We present simulation evidence i
 llustrating our theoretical results and the good properties of the spectra
 l algorithm.\n
DTSTART:20200313T193000Z
DTEND:20200313T210000Z
LOCATION:Leacock 429\, Leacock Building\, CA\, QC\, Montreal\, H3A 2T7\, 85
 5 rue Sherbrooke Ouest
SUMMARY:Mathieu Marcoux (Université de Montréal)\, “Imposing equilibrium re
 strictions in the estimation of dynamic discrete games'
URL:/economics/channels/event/mathieu-marcoux-universi
 te-de-montreal-imposing-equilibrium-restrictions-estimation-dynamic-discre
 te-301813
END:VEVENT
END:VCALENDAR