s'authentifier
rss feed
HAL : halshs-00461711, version 1

Fiche détaillée  Récupérer au format
A short note on the nowcasting and the forecasting of Euro-area GDP using non-parametric techniques
Dominique Guegan 1, 2, Patrick Rakotomarolahy 1
(01/2010)

The aim of this paper is to introduce a new methodology to forecast the monthly economic indicators used in the Gross Domestic Product (GDP) modelling in order to improve the forecasting accuracy. Our approach is based on multivariate k-nearest neighbors method and radial basis function method for which we provide new theoretical results. We apply these two methods to compute the quarter GDP on the Euro-zone, comparing our approach, with GDP obtained when we estimate the monthly indicators with a linear model, which is often used as a benchmark.
1 :  Centre d'économie de la Sorbonne (CES)
CNRS : UMR8174 – Université Paris I - Panthéon Sorbonne
2 :  Ecole d'Économie de Paris - Paris School of Economics (EEP-PSE)
Ecole d'Économie de Paris
Axe Finance
Sciences de l'Homme et Société/Economie et finances

Sciences de l'Homme et Société/Gestion et management

Sciences de l'Homme et Société/Méthodes et statistiques

Mathématiques/Probabilités

Mathématiques/Statistiques

Statistiques/Théorie
k-nearest neighbors method – radial basis function method – non-parametric – forecasts – GDP – Euro-area.
Liste des fichiers attachés à ce document : 
PDF
10013.pdf(562.7 KB)