Predikce příjmů obecních rozpočtů

The core theme of this bachelor thesis is the prediction of revenues of municipal budgets. For this work I have chosen budgets from the town Humpolec. The main goal of this bachelor thesis is to describe how Humpolec prepares the revenue side of the budget and evaluate which one of the methods of prediction is suitable. These methods of prediction are: simple moving average, exponential smoothing, transformation moving average, regression against time and expert's assessment of the town Humpolec. During the calculation of prediction I determined that the expert's assessment is not the most ideal prediction method for all tax revenues. The expert's assessment is the right prediction method for all types of the individual income tax and for the corporation income tax paid by municipalities. For the value added tax and corporation income tax it would be more suitable use the regression against time method. Transformation moving average is the best for prediction of fees. It is more complicated with real estate tax, because no method of prediction emerged as a definitive "winner". For this kind of tax revenue I would choose either the expert's assessment or the regression against time method. Through my research I have demonstrated that it is not recommended to use one method of prediction for all revenues; rather, it is better to assess each section and choose the method of prediction best suited to the specific revenue type.