Abstract: In order to test whether the assumption of homoscedasticity is valid or not for Rwanda Corporate Income tax, we estimated a correctly specified Autoregressive Integrated Moving Average (ARIMA) model of the underlying time series, this helped to remove the linear dependence in the series. To test for Autoregressive Conditional Heteroscedasticity (ARCH) effects the residuals of the mean equation from the ARIMA model have been used. It was verified through the Autocorrelation Function (ACF), Partial Autocorrelation Function (PACF) and Ljung-Box Q-statistics that the residuals of the mean equation did not show any significant serial correlation and that the model provided a good fit. The most appropriate specification model the volatility of Corporate Income tax collections in Rwanda was found to be a GARCH (1,1) model given the characteristics of the time series and the overall fit of the model. The autocorrelation functions of the residuals and squared residuals were also examined to confirm the model adequacy.
Keywords: Volatility, Corporate income tax, Autoregressive, and Heteroscedasticity.
Title: Volatility Modelling of Corporate Income Tax (CIT) in Rwanda
Author: Lucie NIYIGENA, Dr. Marcel Ndengo, Dr. Joseph K. Mung’atu
International Journal of Thesis Projects and Dissertations (IJTPD)
Research Publish Journals
