Modelos de previsão como ferramenta para o planejamento na gestão municipal: uma aplicação para a Prefeitura de Maracanaú-Ce

Abstract

This paper aims to offer a tool for planning and management directed towards the development of related management activities of the Public Budget. The study is a bibliographic and quantitative character, which was used variables of income and expenditure of Ceará Maracanaú comprising the period between 2010 and 2013. Methodology Time Series, in particular, single equation autoregressive models (ARIMA) and autoregressive vector (VAR) was applied, for performing prediction of the main items of expenditure and revenue of the municipality analyzed, and the performance of the predictive models calibrated according to the minimization of the square root of the Mean Square Error (MSE) for the last year for which data were available and the forecast of the items was held by the end of 2014. According to the results, we realized the superiority of ARIMA models relative to VAR models for variables Tax Revenue, Expenditure on Education, Health Expenditure and Expenses currents, since the vector proposal got more grip in the modeling and prediction of Current Revenue Public Administration of Maracanaú. Despite the superiority of one method over another, it is underlined the importance of finding time series as planning mechanisms of the public budget in any sphere of government tools.