BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//jEvents 2.0 for Joomla//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:7ae6bce206e73c943f0bf02ca2de2dbf
CATEGORIES:Seminars
CREATED:20150211T181805
SUMMARY:Lunch Seminar: Marco Lippi - EIEF
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:Forecasting US Inflation with a Dynamic Factor Model\nAbstract:\nThe paper 
 studies the pseudo real-time forecasting performance of three different fac
 tor models. We compare the method recently proposed by Forni et al. (2015) 
 and Forni et al. (2014) with those proposed in Forni et al. (2005) and Stoc
 k and Watson (2002a) within a real data forecasting exercise. A large panel
  of macroeconomic and financial time series for the US economy which includ
 es the Great Recession and the subsequent recovery is employed. In a rollin
 g window framework, we find that the first two methods, based on spectral e
 stimation, outperform the third. Substantial gains from regularized combina
 tions of different inflation forecasts produced with the model in Forni et 
 al. (2015) are also found.\n
DTSTAMP:20260404T122101Z
DTSTART:20141209T130000Z
DTEND:20141209T140000Z
SEQUENCE:0
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR