Hypernormal Densities

Raffaella Giacomini (), Andreas Gottschling, Christian Haefke () and Halbert White
Additional contact information
Andreas Gottschling: Deutsche Bank

No 584, Boston College Working Papers in Economics from Boston College Department of Economics

Abstract: We derive a new family of probability densities that have the property of closed-form integrability. This flexible family finds a variety of applications, of which we illustrate density forecasting from models of the AR-ARCH class for U.S. inflation. We find that the hypernormal distribution for the model's disturbances leads to better density forecasts than the ones produced under the assumption that the disturbances are Normal or Student's t.

Keywords: ARMA-GARCH models; neural networks; nonparametric density estimation; forecast accuracy (search for similar items in EconPapers)
JEL-codes: C63 C53 C45 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-ecm, nep-ets and nep-mfd
Date: 2002-09-01

Downloads: (external link)
http://fmwww.bc.edu/EC-P/WP584.pdf main text (application/pdf)

Related works:
Working Paper: Hypernormal Densities (2002)
Working Paper: Hypernormal Densities (2002)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: http://EconPapers.repec.org/RePEc:boc:bocoec:584

Access Statistics for this paper

More papers in Boston College Working Papers in Economics from Boston College Department of Economics
Address: Boston College, 140 Commonwealth Avenue, Chestnut Hill MA 02467 USA
Contact information at EDIRC.
Series data maintained by Christopher F Baum ().