Simultaneous analysis of revenues from multiple government agencies
Revista do TCU · 2026
DOI: 10.69518/2186 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista do TCU, 2026. Verfasst von Otoniel Arruda Costa und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Budget planning is legally mandated, yet many administrators still face challenges in revenue forecasting. Despite advances in data science, gaps remain in the application of predictive models for this purpose. This study proposes a computational model in R capable of simultaneously analyzing time series of budget revenues from multiple collecting agencies, automatically comparing different methods – Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing. The developed algorithm enables analysts to review outputs from all methods using a single block of code, eliminating the need for repetitive function calls. The algorithm was developed using data from the portals of the Sao Paulo State Court of Accounts (TCE-SP) and the State System Foundation for Data Analysis (SEADE). Two cities with similar economic characteristics as of December 31, 2020 – Sorocaba and Sao Jose dos Campos – were selected, and time series forecasting models were generated for their revenues for the 2022 and 2023 fiscal years. The methodology involved comparing models using the Akaike Information Criterion (AIC), the Corrected Akaike Information Criterion (AICc), and the Bayesian Information Criterion (BIC), alongside the Mean Absolute Percentage Error (MAPE) statistic. The ARIMA method yielded the best information criterion results for both cities and the lowest MAPE for Sao Jose dos Campos (9.75%). With the Exponential Smoothing method, Sorocaba showed the lowest MAPE (12.31%). The results demonstrate the viability of the proposed model in facilitating comparative analyses between municipalities and supporting decision-making by public administrators, auditors, and other stakeholders interested in government economic data.
Zitieren
APA
Costa, O. A., & Góia, H. R. (2026). Simultaneous analysis of revenues from multiple government agencies. Revista do TCU. https://doi.org/10.69518/2186
Vancouver
Costa OA, Góia HR. Simultaneous analysis of revenues from multiple government agencies. Revista do TCU. 2026. doi:10.69518/2186
BibTeX
@article{Costa2026,
author = {Costa, Otoniel Arruda and Góia, Henrique Raymundo},
title = {Simultaneous analysis of revenues from multiple government agencies},
journal = {Revista do TCU},
year = {2026},
doi = {10.69518/2186}
}
EUVIMED