PREDICTION OF TAX REVENUES WITH MULTIVARIABLE GREY PREDICTION MODEL UNDER THE SCOPE OF PUBLIC FINANCE MANAGEMENT

Public finance management is one of the main elements of the reform that was implemented after 1980 in Turkey. The objectives such as accountability, transparency, governance, and fiscal discipline in public activities are envisaged with the public finance management system. In the new system, accurate estimations of income have become important in terms of guiding policymakers. In this study, a new multivariable grey prediction model was used to predict tax revenues in Turkey. This model, which is a model of the grey system theory, enables prediction with many variables without the need for statistical assumptions. In the sample set of this model, tax revenue was used as dependent variable, income taxes, property taxes, domestic taxes on goods and services, and taxes on foreign trade were used as independent variables. The data for 2006-2016 period were evaluated in the simulation of the model, and 2017-2019 period were evaluated in the prediction of the model. According to the experimental results, it was seen that both the simulation errors and the prediction errors of the grey prediction model were quite low. It is believed that these results will provide the basis for a meaningful and effective model for predicting tax revenues.