African Journal of Applied Statistics

Intervention time series modeling with parametric and nonparametric approach: comparative study on Corporation tax in Togo

Dodema BITENIWE and Kossi Essona GNEYOU

Full-text: Access denied (subscription has expired)

We're sorry, but we are unable to provide you with the full text of this article because we are not able to identify you as a subscriber. If you have a personal subscription to this journal, then please login. If you are already logged in, then you may need to update your profile to register your subscription. Read more about accessing full-text


Time series are often subject to structural changes caused by external events such as strikes, new fiscal measures, or policy changes. In the current paper, we conduct a comparative study of a parametric and nonparametric approach to intervention time series modeling to model the impact, on corporate tax, of the important tax reform in December 2012 in Togo (the establishment of a Togolese Revenue Office).The comparison of the two models has led us to conclude that the non-parametric approach is superior in terms of predictive quality as well as the measurement of the effect of the reform.


Les séries temporelles sont souvent sujettes à des changements structurels causés par des événements externes tels que des grèves, de nouvelles mesures fiscales ou des changements de politique. Dans cet article, nous menons une étude comparative des approches paramétrique et non-paramétrique de l'analyse d'intervention afin de modéliser l'impact de l'importante réforme fiscale de décembre 2012 au Togo (mise en place de l'Office Togolais des Recettes) sur l'impôt sur les sociétés. La comparaison des deux modèles nous a amené à conclure que l'approche non-paramétrique est meilleure en terme de qualité prédictive ainsi que la mesure de l'effet de la réforme.

Article information

Afr. J. Appl. Stat., Volume 6, Number 2 (2019), 711-722.

First available in Project Euclid: 12 November 2019

Permanent link to this document

Digital Object Identifier

Primary: 62P20: Applications to economics [See also 91Bxx]
Secondary: 62M10: Time series, auto-correlation, regression, etc. [See also 91B84] 91B84: Economic time series analysis [See also 62M10] 62G08: Nonparametric regression

intervention analysis central mean subspace in time series NadarayaWatson kernel estimator


BITENIWE, Dodema; GNEYOU, Kossi Essona. Intervention time series modeling with parametric and nonparametric approach: comparative study on Corporation tax in Togo. Afr. J. Appl. Stat. 6 (2019), no. 2, 711--722. doi:10.16929/ajas/2019.711.238.

Export citation