An exploratory hybrid AI workflow for Brazilian federal budget allocation
Data & Policy · 2026 · 8
DOI: 10.1017/dap.2026.10062 ↗Kurz gefasst
Fachartikel in der Zeitschrift Data & Policy, online erschienen am 08.04.2026. Verfasst von Saulo de Oliveira Nonato und 4 weiteren Autor:innen. Die Arbeit nennt 70 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study assesses whether a hybrid prediction–optimisation workflow can be used as an exploratory exercise for Brazilian federal budget allocation under severe data constraints. Using executed expenditure by budgetary function (2000–2023; N = 24), a multi-output XGBoost model is estimated to link spending profiles to GDP growth, inflation, and the Gini index; Bayesian optimisation (Tree-structured Parzen Estimator/Optuna) is then applied to search, within explicit bounds and penalties, for allocation vectors that maximise a stated objective function favouring higher growth and lower inflation and inequality. To mitigate data scarcity, the short series is augmented with 1048 synthetic observations generated through controlled noise injection, bootstrapped resampling and variational autoencoder reconstruction. Under randomised K-fold cross-validation on the augmented dataset, the model achieves mean R 2 = 0.97 and mean MSE = 0.04, while diagnostics indicate larger errors at extreme values and a persistent training–validation gap. A secondary robustness check uses an anti-leakage design by applying cross-validation to the 24 real observations and generating synthetic data only within each training fold. This yields markedly weaker generalisation for GDP growth and inflation (overall mean MSE = 1.03; overall mean R 2 = −0.45), with positive performance remaining only for the Gini index ( R 2 = 0.60). Under these conditions, the optimisation step identifies a scenario that satisfies the objective function on standardised outputs (GDP growth = 1.15; inflation = −0.04; Gini = −0.17). The results support the use of the workflow to compare scenarios under explicit assumptions, rather than to produce prescriptive budget guidance.
Zitieren
APA
Nonato, S. D. O., Moreira, M. F., Prata, D. N., Lima, D. V. D., & Prata, G. R. N. (2026). An exploratory hybrid AI workflow for Brazilian federal budget allocation. Data & Policy, 8. https://doi.org/10.1017/dap.2026.10062
Vancouver
Nonato SDO, Moreira MF, Prata DN, Lima DVD, Prata GRN. An exploratory hybrid AI workflow for Brazilian federal budget allocation. Data & Policy. 2026;8. doi:10.1017/dap.2026.10062
BibTeX
@article{Nonato2026,
author = {Nonato, Saulo de Oliveira and Moreira, Marina Figueiredo and Prata, David Nadler and Lima, Diana Vaz de and Prata, Gabriel Reis Nadler},
title = {An exploratory hybrid AI workflow for Brazilian federal budget allocation},
journal = {Data & Policy},
year = {2026},
volume = {8},
doi = {10.1017/dap.2026.10062}
}
EUVIMED