Um modelo de previsão para o planejamento da programação financeira da Marinha do Brasil

Abstract

Efficient management of public financial resources represents a continuous challenge for public managers, being fundamental to achieving high standards of effectiveness in the services provided to society. In this scenario, the financial sector of the Brazilian Navy (MB) is tasked with planning, distributing, and controlling the financial resources necessary to meet the demands of all the force's managing units. In this context, statistical forecasting methods emerge as a valuable tool to be employed in the planning and control of public organizations. With the aim of improving the accuracy of financial planning for the Brazilian Navy (MB), this study seeks to evaluate the performance of statistical forecasting methods for planning the MB's Financial Programming. A descriptive quantitative methodology was adopted, analyzing monthly payment data from 2017 to 2022. The efficacy of the models was assessed by comparing the forecasts made by the additive and multiplicative Holt-Winters methods and by ARIMA with the actual figures for 2022. Subsequently, projections for the year 2023 were generated. The results indicated that the ARIMA method had the best performance, generating an annual forecast for the total value of financial resource needs with a relative deviation of just 7% compared to the amount executed in 2022 and 10.9% for 2023. These results demonstrate the significant potential of statistical forecasting methods as support tools for financial planning in the MB, contributing to a more strategic and effective management.