Application Des Modeles Binomiaux Negatifs De Type P (nb-p) Et De Type 2 (nb2) a La Modelisation Du Risque De Fluctuation Des Recettes Fiscales : Cas De La Direction Generale Des Impots En Republique Democratique Du Congo
Year: 2026
Type: article
Abstract: A balanced tax revenue stream is crucial for managing public finances, particularly in the Democratic Republic of the Congo (DRC), where tax revenues are vulnerable to economic fluctuations. The objective of this article is to model and forecast the risk of fluctuations in tax revenues collected by the General Tax Directorate (DGI), using an actuarial method based on negative binomial models.First, negative binomial models (Type P and Type 2) are used to model the risk of fluctuations in tax revenue in general. Next, the resulting model is applied to the case of the DGI in the DRC. The study is based on historical monthly tax data from 2014 to 2023, representing 120 observations. A comparative analysis is then conducted between the NB-P model and the negative binomial Type 2 (NB2) model to determine the specification best suited to the data.The results highlight an overdispersion of fiscal fluctuations, with above-average variance. The NB-P model performs better than the NB2 model, with a higher log-likelihood (-245.63 versus -252.84), as well as lower AIC (503.26) and BIC (524.87) values. The estimated power parameter of 1.350 confirms that the variance structure is not strictly quadratic. The results also Ashow that GDP and the number of taxpayers reduce the risk of volatility, while inflation and the exchange rate increase
Source: Journal Africain des Sciences
Institution: University of Kinshasa
Language: English
Topic: Economic and Fiscal Studies
Subfield: Economics and Econometrics
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7221338838
