Paper #638

Títol:

Hypernormal densities

Autors:

Raffaella Giacomini, Andreas Gottschling, Christian Haefke i Halbert White

Data:

Setembre 2002

Resum:

We propose a new family of density functions that possess both flexibility and closed form expressions for moments and anti-derivatives, making them particularly appealing for applications. We illustrate its usefulness by applying our new family to obtain density forecasts of U.S. inflation. Our methods generate forecasts that improve on standard methods based on AR-ARCH models relying on normal or Student's t-distributional assumptions.

Paraules clau:

ARMA-GARCH models, neural networks, nonparametric density estimation, forecast accuracy

Codis JEL:

C63, C53, C45

Àrea de Recerca:

Estadística, Econometria i Mètodes Quantitatius

Descarregar el paper en format PDF