Statistics Modeling and Predicting Random Effects of Tax Revenues Panel Data in DRC: Case of North-Kivu Province

Abstract

This paper aims to model and predict random effects of taxes revenues panel data in DRC. After aggregating data according to weight and Goods, three models have been developed and we found that: The model is not identical to all countries as the effects of the pooled model were non-significant and data have been esteemed by the random effects model because Hausman was non-significant. The dynamic model is not esteemed because of the equilibrium of our panel under study. According to goods as individuals, pooled Model was not significant while fixed individual effects were significant since the least squares estimator in OLS regression of Y on X in the N dummy variable.