Pronóstico del IVA de grandes contribuyentes mediante un modelo estacionario lineal (Tema Central)

In Ecuador, VAT incurred tax collection exceeds 50 % of total tax collection, with large taxpayers make up the majority of tax collection by economic sector, so that modeling and forecasting the evolution of the tax variable is relevant for the formulation of public policies and state budget planning. This research aims to model the VAT incurred tax collection of large taxpayers in the period 2011-2023 through linear stationary models, such as the ARIMA and SARIMA processes, using the Box-Jenkins methodology and informational criteria for the estimation and validation of the model; in addition, to forecast the evolution of this tax variable towards 2024. Once the appropriate model with a practically perfect fit has been identified —this being a SARIMA (1,1,1) X (1,0,1)12—, the forecast for 2024 indicates that collection will increase by 4.73 % with respect to 2023 in December, considering some possible causes of this behavior. In 2024, a sort of collection cycle is maintained with peaks in the last month, while in the others the activity slows down, even in June. A limitation of this research is the non-consideration of GARCH effects and control variables within the estimation, so it is recommended that future research consider these guidelines.