What question did this study set out to answer?

The research aims to assess the feasibility of a hybrid prediction-optimization workflow for allocating the Brazilian federal budget under data constraints.

April 10, 2026Data & PolicyOpen Access

An exploratory hybrid AI workflow for Brazilian federal budget allocation

View Full Paper

Ask AI

Bookmark

Authors

SNSaulo De Oliveira NonatoFilm Independent MMMarina Figueiredo MoreiraUniversidade de Brasília DPDavid Nadler PrataUniversidade Federal do Tocantins

Discussion

Member takes

Implication

Exploratory analysis uncovers budget allocation strategies in Brazil, indicating potential for improved outcomes.

Key Points

The research aims to assess the feasibility of a hybrid prediction-optimization workflow for allocating the Brazilian federal budget under data constraints.
Developed a multi-output XGBoost model linking spending profiles to GDP growth, inflation, and Gini index.
Applied Bayesian optimisation with Tree-structured Parzen Estimator/Optuna to identify optimal budget allocations.
Augmented data by generating 1048 synthetic observations to address data scarcity.
Utilized randomized K-fold cross-validation for model evaluation.
Achieved mean R2 of 0.97 and mean MSE of 0.04 with the augmented dataset.
Found weaker generalization on real observations, with overall mean MSE of 1.03 and mean R2 of -0.45.
Identified a scenario that maximizes the objective function with GDP growth at 1.15, inflation at -0.04, and Gini at -0.17.

Cite This Study

Nonato et al. (2026) studied this question.

synapsesocial.com/papers/69d895be6c1944d70ce06d99 https://doi.org/10.1017/dap.2026.10062

View Full Paper

Ask AI

Bookmark

An exploratory hybrid AI workflow for Brazilian federal budget allocation

Discussion

Member takes

Implication

Key Points

Cite This Study

An exploratory hybrid AI workflow for Brazilian federal budget allocation

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider