A Short Note on the Nowcasting and the Forecasting of Euro-area GDP Using Non-Parametric Techniques
Article dans une revue: 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.
Auteur(s)
Dominique Guegan, Patrick Rakotomarolahy
Revue
- Economics Bulletin
Date de publication
- 2010
Mots-clés
- Multivariate k-Nearest Neighbor
- Radial Basis Functions
- Non-Parametric Forecasts
- Economic indicators
- GDP
- Euro area
Pages
- 508-518
URL de la notice HAL
Version
- 1
Volume
- 30