La prévision de l'inflation par la méthode des réseaux de neurones: cas de la Tunisie

Damien Bazin, Inès Abdelkafi () and Rochdi Feki ()
Additional contact information
Inès Abdelkafi: Unité de Recherche en Economie du Développement (URED) - Chercheur indépendant
Rochdi Feki: Unité de Recherche en Economie du Développement (URED). - Chercheur indépendant

Post-Print from HAL

Abstract: The neural approach drew the interest of many researchers for time series analysis and forecasting in diverse domains. In this paper, we study the ability of artificial neural networks (ANN) such as "multilayer perceptrons" to predict the Tunisian inflation rate. We try to find a better technical of inflation forecasting by comparing the results obtained using ANN to those provided by linear autoregressive models (AR) and the "naive" forecasting model. The comparison is based on the root-mean-square error (RMSE) criterion and the improvement rate of the latter (measured against the random walk). The results found showed the superiority of the RNA to trace the series evolution and to offer a better performance in terms of predictive power for inflation rate in Tunisia.

Keywords: Inflation rate; forecasting time series; artificial neural networks.; prévision des séries temporelles; réseaux de neurones artificiels.; réseaux de neurones artificiels; Taux d'inflation (search for similar items in EconPapers)
Date: 2012-06-14
References: Add references at CitEc
Citations:

Published in Éthique et économique/Ethics and economics, 2012, 9 (1), pp.141-151

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
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: https://EconPapers.repec.org/RePEc:hal:journl:halshs-00727319

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().