Prediction of dengue cases based on climatic variables by machine learning

doi:10.22167/2675-6528-202602684

Article

September 30, 2026

Prediction of dengue cases based on climatic variables by machine learning

Prediction of Dengue Cases Based on Climatic Variables by Machine Learning

Anna Beatriz Pereira Lima; Brayan Dias Dauria

DOI: 10.22167/2675-6528-202602684

Article derived from a Course Conclusion Work (TCC), with content based on the student’s original work and adapted to the editorial format of the E&S Magazine with the support of the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege for textual synthesis and organization.

Summary

Dengue has characterized itself as an important public health problem in Brazil, with its occurrence influenced by climatic, environmental, and temporal factors. The study aimed to develop and evaluate predictive models to estimate the number of dengue cases in the state of São Paulo by integrating epidemiological data from the Information System for Notifiable Diseases (SINAN) and meteorological variables from the National Institute of Meteorology (INMET). The investigation was conducted as a case study, based on a temporal structure at the municipality-month level, built from the processing, standardization, and integration of databases. The methodological process included the selection and aggregation of confirmed notifications, the spatial association between municipalities and meteorological stations, the creation of lagged and moving average variables, as well as the temporal division of data into training, validation, and testing sets. The Random Forest, Support Vector Regression, XGBoost, and Long Short-Term Memory models were applied, evaluated using MAE, RMSE, and coefficient of determination (R²) metrics. The results showed that the models captured relevant patterns from the historical series, especially the seasonality of the disease and the influence of the recent history of cases and climatic variables. Among the evaluated algorithms, Random Forest presented the best overall performance, demonstrating potential to support epidemiological surveillance, monitoring, and outbreak anticipation actions.

Keywords: Dengue; Epidemiological forecasting; Public health; Time series; Climate variables.

1. Introduction

Dengue is an arbovirus disease caused by the DENV virus, transmitted to humans by the bite of an infected Aedes aegypti mosquito (World Health Organization, 2025). The virus belongs to the Flaviviridae family and the Orthoflavivirus genus, presenting four known serotypes: DENV-1, DENV-2, DENV-3, and DENV-4, which differ by genetic variations and specific lineages (Cardoso et al., 2024). Widely disseminated in tropical and subtropical regions, the disease represents a serious public health problem in Brazil, one of the countries with the highest number of cases and recurrence of epidemics (Felix, 2024).

The spread of dengue is directly related to environmental and climatic conditions that favor the mosquito’s life cycle and the virus’s dissemination. Aedes aegypti, the main transmitter of the disease, finds ideal conditions for reproduction in urban areas, especially in places with stagnant water accumulation and precarious sanitation (Carvalho Da Silva et al., 2024). Variations in temperature, humidity, and precipitation directly influence the mosquito’s life cycle and viral replication, making climate a determining element in the disease’s dynamics (Gui et al., 2021). Thus, the interaction between biological, environmental, and social factors contributes to the maintenance and intensification of dengue cases in the country.

Faced with a large, dense, and heterogeneous population, Brazil faces significant challenges related to the increase in dengue cases, the occurrence of which is strongly influenced by socioeconomic and seasonal factors. The severity of this problem becomes even more evident, with the registration of more than a thousand deaths from dengue in 2025, according to data from the Ministry of Health (Coelho, 2025). Social inequality contributes to precarious housing, sanitation, and waste management conditions, creating environments conducive to the proliferation of the mosquito (Bezerra, T.; Matos, C. 2023).

In this context, hybrid models based on machine learning and deep learning show promise in predicting dengue cases, using climatic and epidemiological data (Lanlan, F., 2024). Studies in the field have applied machine learning algorithms such as Random Forest, Extreme Gradient Boosting, and Support Vector Machine/Regression, recognized for their ability to capture complex relationships between multiple climatic factors and the dynamics of dengue case variation (Ong et al., 2023). Within the scope of deep learning, recurrent neural networks of the Long Short-Term Memory type have also shown expressive results in predicting dengue based on climatic variables (Xiang, C.; Morgan, P., 2025; Lanlan, F. et al., 2024).

Although the main focus of this study is the prediction of dengue cases, the proposed methodological approach has the potential for extension to other infectious diseases influenced by environmental and climatic factors, such as chikungunya and zika. The integrated use of epidemiological data and climatic variables, combined with machine learning techniques, allows for the construction of predictive models applicable to different public health contexts. This versatility expands the potential application of the solution, enabling its use in various geographical regions and contributing to the development of national-scale monitoring and alert systems. Thus, the objective of this work was to develop and compare machine learning and deep learning models applied to the prediction of dengue cases from climatic variables and historical epidemiological data, evaluating the performance of different algorithms in capturing temporal patterns and nonlinear relationships associated with the disease dynamics, as well as investigating their potential as a tool to support epidemiological surveillance, contributing to the anticipation of outbreaks and to public health decision-making.

2. Material and Methods

The research developed was characterized as a study of an applied and descriptive nature. The applied approach aimed at solving a practical problem, focused on predicting dengue cases from climatic variables and epidemiological data, with potential to support public health surveillance. Regarding the objectives, the research was descriptive, seeking to identify and organize temporal patterns, as well as relationships between environmental variables and the incidence of the disease, without establishing direct causal relationships.

The methodological procedure consisted of a case study, in which different machine learning and deep learning models were applied to the same dataset to compare their performances. The study steps were organized into five main phases, which included data acquisition, database processing, information integration, algorithm modeling and implementation, and analysis of the obtained results.

For the analysis, epidemiological data of dengue cases that occurred in Brazil and associated meteorological information were used. The epidemiological data were obtained from the Information System for Notifiable Diseases (SINAN), made available by the Informatics Department of the Unified Health System (DATASUS), covering the period from 2014 to 2025. The meteorological data came from the National Institute of Meteorology (INMET), referring to the municipalities of the state of São Paulo.

The collection of epidemiological data from SINAN was carried out through a connection via File Transfer Protocol (FTP), using FileZilla software. The databases, originally in DataBase Compressed (DBC) format, were converted to DataBase File (DBF) with TabWin software and, subsequently, to Comma-separated values (CSV) via Python language. The columns DT_NOTIFIC, ID_MUNICIP, FEBRE, CEFALEIA, VOMITO, RESUL_SORO, RESUL_NS1, RESUL_PCR_, HOSPITALIZ, CLASSI_FIN, and CRITERIO were selected, converting the notification date to temporal format and standardizing municipal codes.

SINAN data were filtered to include only cases from the state of São Paulo, selecting notifications with a final dengue classification (codes 10, 11, or 12). Subsequently, the data were aggregated monthly by municipality, using the IBGE code as an identifier. A complete temporal structure was created for all municipality, year, and month combinations, filling months with no confirmed cases with zero.

INMET meteorological data were obtained through automated email submission, generating an individual file for each station. Climate information, collected on a monthly scale, included variables such as accumulated precipitation, average air temperature, and average wind speed. The consolidation of these databases was carried out by an automated process implemented in Python.

To associate the climatic data with the municipalities, it was considered that the state of São Paulo has 645 municipalities and the INMET network has 43 meteorological stations. Each municipality was associated with the nearest meteorological station. The municipal grid of IBGE in shapefile format was used, representing the stations as geographic points and the municipalities by their centroids. The linkage was carried out based on the spatial proximity between the municipal centroids and the meteorological station points, a procedure similar to that adopted by Aune et al. (2023) and Gómez-Barroso et al. (2017).

After data processing, epidemiological and meteorological data were joined by municipality, year, and month. Lagged variables, known as “lags”, were created to represent the influence of values observed in previous periods on the number of dengue cases. For the target variable confirmed_cases, lags of one, two, three, six, and twelve months were used. Lags were also created for climatic variables, and moving averages with temporal displacement were calculated to avoid data leakage.

The consolidated dataset was divided into training, validation, and test sets, temporally, without random shuffling of records, in order to preserve the chronological order of observations. The year 2025 was reserved as the final test set, while the preceding years were used in the training and validation stages. The month variable was transformed into cyclic sine and cosine components to incorporate the seasonal nature of the calendar.

In the training base, temporal validation was applied using the “TimeSeriesSplit” method, with five divisions. This procedure partitioned the data into successive divisions over time, keeping the older observations for training and the later ones for validation. In each iteration, the model was trained with a historical period and validated on a subsequent period, simulating a realistic forecasting scenario.

For the definition of the models’ hyperparameters, the grid search technique was used through the “GridSearchCV” method. This procedure systematically evaluated different combinations of hyperparameters, using the temporal splits defined by “TimeSeriesSplit”. The selection of the best configurations was carried out exclusively on the training and validation data, considering the performance on the R², MAE, and RMSE metrics.

In the predictive modeling stage, the Random Forest (RF), Support Vector Regression (SVR), XGBoost, and Long Short-Term Memory (LSTM) models were applied. Random Forest, a supervised learning method, combined multiple decision trees to estimate continuous values, using the bagging technique (Rahman et al., 2025). XGBoost, based on boosting, trained decision trees sequentially to correct previous errors (Ribeiro et al., 2021).

Support Vector Regression (SVR), implemented through the LinearSVR algorithm, sought to estimate a continuous function with error limited by a tolerance margin (Leão, 2025). The Long Short-Term Memory (LSTM) neural network was used as a recurrent neural network architecture, designed to capture short- and long-term temporal dependencies in time series (Chen et al., 2025). For the XGBoost and LSTM models, the log1p transformation was applied to the target variable to reduce the asymmetry of the distribution and mitigate the influence of extreme values.

The research development was carried out in a computational environment configured for data treatment, integration, analysis, and modeling. The programming language used was Python, due to its versatility and the wide availability of libraries. Libraries such as Pandas for data manipulation, NumPy for numerical operations, Scikit-learn for preprocessing and evaluation, XGBoost and TensorFlow/Keras for model implementation, Matplotlib for visualization, and GeoPandas for spatial data manipulation were employed.

The evaluation of the models constituted a fundamental step, carried out based on the predictive performance obtained on the final test set, corresponding to the year 2025. For this comparison, the metrics Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R²) were used. The MAE measured the average of the absolute differences between the actual and predicted values, while the RMSE quantified the model’s accuracy relative to the actual values, calculating the square root of the average of the squared errors. The R² represented the percentage of the data variance explained by the model.

3. Results and Discussion

This section details the results obtained from applying the previously described methodological steps, covering data collection, processing, analysis, and merging of databases, as well as predictive modeling and algorithm evaluation. The main objective was to demonstrate the progress achieved in predicting dengue cases in the state of São Paulo, identify the particularities observed during the process, and discuss the implications of these findings for epidemiological surveillance. The systematic approach allowed for an in-depth understanding of the disease dynamics and the models’ capacity to capture complex patterns.

Database and Pre-processing

The construction of the database for this study involved the integration of epidemiological and meteorological information, fundamental for the predictive modeling of dengue. Dengue case data were extracted from the Notifiable Diseases Information System (SINAN), covering the period from 2014 to 2025. Initially, this data, available in DBC format, was converted to DBF and subsequently to CSV, using tools such as FileZilla, TabWin, and the Python language, ensuring compatibility for analysis. This conversion step was crucial for standardizing the information and enabling its manipulation in the computational environment.

For the treatment of epidemiological data, a rigorous filtering was performed to include only confirmed dengue cases in the state of São Paulo, identified by final classification codes (CLASSI_FIN) 10, 11, or 12. Notifications, originally daily, were aggregated monthly by municipality, using the IBGE code as an identifier. Clinical and laboratory variables, such as fever, headache, vomiting, and examination results (NS1, RT-PCR, serology), were also aggregated, providing a detailed overview of the cases. The standardization of municipal codes and the filling of months with no records with zero cases were essential to create a balanced and complete time series, ready for integration.

The meteorological data were obtained from the National Institute of Meteorology (INMET), referring to stations located in the state of São Paulo. The variables included accumulated precipitation, average air temperature, and average wind speed, collected on a monthly scale. A significant challenge was the disparity between the number of municipalities (645) and meteorological stations (43) in the state. To address this issue, each municipality was associated with the nearest meteorological station, using the municipal grid from IBGE in “shapefile” format and calculating the spatial proximity between municipal centroids and station points, following approaches by Aune et al. (2023) and Gómez-Barroso et al. (2017).

The spatial association resulted in a mapping table that linked the 645 municipalities to the 43 INMET stations, with an average distance of 33.01 km and a maximum distance of 152.75 km between municipal centroids and stations. This step was fundamental to incorporate climatic variables into the epidemiological database, allowing for a joint analysis. The final consolidated database totaled 92,880 records, covering 645 municipalities over 12 years, from 2014 to 2025. The month variable was transformed into cyclic sine and cosine components to capture annual seasonality more precisely, recognizing the circular nature of the calendar.

The exploratory analysis of the aggregated time series of dengue cases in the state of São Paulo revealed significant oscillations and recurrent epidemic peaks. Among the years evaluated, 2024 stood out as the period with the highest volume of cases, configuring the main peak in the historical series. In contrast, years like 2017 and 2018 presented considerably reduced totals, characterizing periods of lower disease circulation. Meanwhile, 2025, although remaining at a high level, presented a total lower than that observed in 2024. The seasonality of the disease was clearly observed, with the highest concentration of cases in the first months of the year, especially in March, April, and May, a pattern that aligns with the vector’s life cycle and favorable climatic conditions.

The climatic variables also exhibited a pronounced seasonal behavior, compatible with the dynamics of dengue. The average monthly precipitation was higher in the first and last months of the year, with a notable reduction between June and August. The average temperature, in turn, remained higher between October and March, decreasing in the central months. This seasonal correlation between climatic patterns and disease incidence reinforced the importance of these variables for predictive modeling, as discussed by Gui et al. (2021) on the influence of climate on the spread of dengue.

However, the analysis of INMET meteorological data also revealed the presence of missing values, with greater incompleteness between 2020 and 2022. Precipitation was the variable with the highest percentages of absence, while average temperature showed a lower proportion of missing data. These limitations in the temporal coverage of the climate database were considered in the subsequent stages of the study, influencing the robustness of the models. The management of this missing data was a critical aspect for maintaining the integrity of the time series and the validity of the forecasts.

To capture the temporal dependence of the series, lagged variables, or “lags”, were created for both the number of confirmed cases and the climatic variables. For the target variable, lags of 1, 2, 3, 6, and 12 months were used, reflecting the influence of previous periods. For the climatic variables, in addition to lags, moving averages with temporal displacement were calculated, ensuring that only past information was used in the forecast, avoiding data leakage and preserving the chronological order of records, an essential procedure in time series (Sacay et al., 2025; Chen et al., 2025).

The autocorrelation analysis of the confirmed cases series showed a strong temporal dependence, with a lag of 1 month presenting the highest correlation (0.822), followed by a gradual decrease. Lags of 11, 12, and 24 months also showed relevant values (0.363, 0.349, and 0.349, respectively), confirming the annual seasonality of the disease. The lagged climatic variables also presented significant correlations, with average temperature at lags of 3 (0.615) and 2 (0.571), and precipitation at lags of 3 (0.514) and 2 (0.488), indicating that the relationship between climate and dengue is not immediate, but time-lagged.

The database was then organized for the predictive modeling stage, with the year 2025 reserved as the final test set, and the preceding years for training and validation. Temporal validation was performed using the “TimeSeriesSplit” method, with five divisions, ensuring that the chronological sequence of the data was maintained. This method simulates a situation closer to the real use of the model, where only past information is available at the time of prediction. Hyperparameter selection was carried out using “GridSearchCV”, optimizing performance based on the R², MAE, and RMSE metrics.

Predictive Modeling and Algorithm Configuration

In the predictive modeling stage, four distinct algorithms were applied: Random Forest (RF), Support Vector Regression (SVR), Extreme Gradient Boosting (XGBoost), and Long Short-Term Memory (LSTM). Each model was configured with hyperparameters optimized through the “GridSearchCV” technique, which sought the best combinations to maximize predictive performance in the context of the dengue cases time series. The choice of these algorithms reflects the diversity of approaches in machine learning and deep learning for time series regression problems.

The Random Forest model, as described by Rahman et al. (2025), was configured with 200 trees, a minimum size of 3 samples per terminal leaf, and 10 minimum samples for splitting nodes, using all available variables in each partition. This architecture, based on the “bagging” technique, is known for its ability to reduce variance and the risk of overfitting, combining multiple decision trees to generate robust predictions. The optimized configuration aimed to enhance the model’s generalization to unseen data.

For Support Vector Regression (SVR), implemented via the “LinearSVR” algorithm, the selected hyperparameters were C=3.0, epsilon=0.05, random_state=42, max_iter=30000, and dual=auto. According to Leão (2025), SVR seeks to estimate a continuous function with limited error, being effective in modeling linear and non-linear relationships, even in high-dimensional scenarios. The optimization of these parameters was fundamental for the model to find a representation that preserved generalization capability and adequately fit the data patterns.

In the XGBoost and LSTM models, a logarithmic transformation (log1p) was applied to the target variable, the number of confirmed dengue cases. The main objective of this procedure was to reduce the asymmetry of the data distribution and mitigate the influence of extreme values, contributing to greater stability in training, especially in a time series characterized by high variability and epidemic peaks. After prediction, the values were converted back to the original scale, allowing for direct interpretation of the results in terms of the actual number of cases.

The XGBoost model, based on “boosting” and described by Ribeiro et al. (2021), was configured with 800 estimators, a learning rate of 0.03, a maximum depth of 4, a minimum child weight of 5, a subsample of 0.85, and a colsample_bytree of 0.85, with reg_lambda of 5.0 and a quadratic error regression objective. This configuration aims to progressively enhance predictive performance by correcting errors from previous trees and incorporating regularization terms to mitigate overfitting, representing a robust approach for regression problems.

For the Long Short-Term Memory (LSTM) neural network, a deep learning architecture suitable for time series, the defined hyperparameters included a window size of 12, 32 units, 0.2 dropout, 16 dense units, Adam optimizer, MSE loss function, 30 epochs, batch size of 64, shuffle FALSE, and an early stopping patience of 5. The LSTM, as per Chen et al. (2025), is designed to capture short- and long-term temporal dependencies, being particularly effective in modeling complex patterns such as trend and seasonality in epidemiological time series.

Performance Evaluation of Models

The performance evaluation of the models was carried out on the final test set, corresponding to the 2025 data, using the Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R²) metrics. The results obtained by the Random Forest, SVR with LinearSVR, XGBoost, and LSTM models, considering the best selected configuration, were: Random Forest (MAE 63.94, RMSE 324.92, R² 0.77); SVR (MAE 55.64, RMSE 360.02, R² 0.72); XGBoost (MAE 86.63, RMSE 551.45, R² 0.34); and LSTM (MAE 51.75, RMSE 435.26, R² 0.59). These metrics allowed for a comprehensive analysis of the accuracy of the predictions and the models’ ability to fit the observed values.

The Random Forest model demonstrated the best overall performance, with an RMSE of 324.93 and an R² of 0.77. Although it did not present the lowest MAE, its balance between predictive accuracy and ability to explain data variability was superior to the others. This result suggests that Random Forest effectively captured the complex relationships between climatic and epidemiological variables and the occurrence of dengue, offering more consistent and reliable predictions for the analyzed period, as observed in similar studies (Rahman et al., 2025).

Support Vector Regression (SVR), using the LinearSVR algorithm, presented the second best performance, with MAE of 55.64, RMSE of 360.03, and R² of 0.72. These values indicate consistent results and relatively close to those of Random Forest, confirming SVR’s ability to model complex relationships and generalize well, even with the presence of noise in the data. The robust performance of SVR positions it as a viable alternative for predicting dengue cases, corroborating findings by Leão (2025).

The Long Short-Term Memory (LSTM) neural network obtained the lowest Mean Absolute Error (MAE) of 51.76, indicating that, on average, its predictions were closest to the actual values in terms of absolute deviation. However, the RMSE of 435.27 and R² of 0.59 suggest lower model stability in the face of more significant errors in certain periods, and a slightly lower capacity to explain data variance compared to Random Forest and SVR. This can be attributed to LSTM’s sensitivity to peaks and abrupt variations in the time series, despite its ability to capture long-term dependencies (Chen et al., 2025).

On the other hand, the XGBoost model presented the least satisfactory performance among the evaluated algorithms, with an MAE of 86.63, RMSE of 551.46, and R² of 0.34. These results indicate a lower capacity to adjust to the observed dynamics of the time series in the test period, suggesting that XGBoost had greater difficulty in handling the complexity and variability of dengue data in the state of São Paulo, especially when compared to the Random Forest and SVR-based approaches, which proved more suitable for this dataset.

The visual analysis of the comparison between the real cases and the monthly forecasts for 2025 revealed that all models were able to capture the general seasonality of the series, characterized by higher values in the first months of the year and a significant reduction in the second half. However, the Random Forest stood out for presenting a behavior closer to the observed trajectory, consistently following the trends of growth and reduction in cases throughout the year. The SVR also reproduced the general shape of the curve, but with a more evident underestimation in the months of highest incidence.

Despite the low MAE, the LSTM exhibited oscillations that distanced it from the observed values in some months of the epidemic period, while XGBoost tended to overestimate cases in peak months, diverging more intensely from the actual series. This visual observation complements the quantitative metrics, reinforcing that Random Forest demonstrated the best ability to reproduce the temporal behavior of the disease. The monthly comparison between actual cases and the values predicted by Random Forest for 2025 indicated that the largest differences were concentrated in the first months of the year, a period of higher disease incidence, while in subsequent months, the predictions came closer to the observed values.

The spatial analysis, through heat maps comparing the actual and predicted cases by Random Forest, indicated a relevant correspondence in the spatial distribution of the disease. Areas with higher intensity of actual cases also showed higher intensity in the model’s predictions, and regions with lower occurrence maintained a similar behavior. This finding suggests that Random Forest was able to adequately represent the spatial heterogeneity of dengue cases in the state of São Paulo, generally preserving the distribution observed among municipalities, which is crucial for epidemiological surveillance.

In summary, the results demonstrate that machine learning and deep learning models are promising tools for predicting dengue cases, with Random Forest standing out as the algorithm with the best overall performance. The models’ ability to capture the seasonality of the disease, the influence of recent case history and climatic variables, as well as spatial heterogeneity, validates the proposed methodological approach. Although limitations related to data availability and quality have been identified, the research offers a significant advancement in understanding dengue dynamics and in developing public health support tools.

4. Conclusion

This study sought to develop and compare machine learning and deep learning models for predicting dengue cases in the state of São Paulo, integrating epidemiological data from SINAN and meteorological variables from INMET. It was found that the models were capable of capturing relevant patterns from the historical series, including the seasonality of the disease and the influence of recent case history and climatic variables. Among the evaluated algorithms, Random Forest demonstrated the best overall performance, with a superior balance between predictive accuracy and the ability to explain data variability, followed by Support Vector Regression. The Long Short-Term Memory neural network presented the lowest mean absolute error, but with less stability regarding more significant errors. Random Forest’s predictions satisfactorily reproduced the temporal and spatial behavior of cases in the analyzed period, highlighting its capacity to represent the heterogeneity of the disease among municipalities.

This approach represents a significant contribution to epidemiological surveillance, offering a potential tool for monitoring and anticipating dengue outbreaks, thereby supporting public health decision-making. However, the study faced limitations related to the difficulty of accessing and standardizing epidemiological databases, the presence of missing values in meteorological data, and the need to associate municipalities with the nearest weather stations. As future perspectives, it is suggested to incorporate new variables, expand the analysis to other geographical regions, and evaluate different forecasting horizons in order to improve the robustness and applicability of the predictive models.

Bibliographic References

Cardoso, R. L.; Corioletti, N. S. D.; Costa, L. F.; Taveira, J. H. da S.; Bueno, C. P.; Corioletti, S.; Giongo, P. R.; Silva, R. T. da. Dengue no Brasil: uma revisão sistemática. Revista Foco, v. 17, n. 3, p. e4640, 2024. DOI: https://doi.org/10.54751/revistafoco.v17n3-079.

Carvalho da Silva, R. et al. Análise dos fatores meteorológicos, socioambientais e socioeconômicos relacionados à incidência de dengue em Joinville – SC nos anos de 2020 e 2021. Hygeia: Revista Brasileira de Geografia Médica e da Saúde, v. 20, p. 1–12, 2024. DOI: 10.14393/Hygeia2071212.

Felix, 2024 [Referência completa não encontrada no documento original]

Gui, H.; Gwee, S.; Koh, J.; Pang, J. Weather factors associated with

World Health Organization, 2025 [Full reference not found in the original document]

To generate the final bibliographic list, I performed data cross-referencing according to the established rules.

Article originating from the Final Course Project of the Specialization in Data Science and Analytics of the MBA USP/Esalq

To learn more about the course, click here and access the MBX Academy platform

Most recent

Economy, Investments And Banking

October 23, 2026

Educação Financeira, Comportamentos e Vieses na Tomada de Decisão Financeira

October 22, 2026

Challenges and Possibilities of the First School Year of Children with ASD for Teaching Work

October 22, 2026

Consumer Journey in the Decision to Purchase Infoproducts in the Digital Environment

October 22, 2026

Psychosocial Risks and Mental Health at Work in a Banking Institution

October 22, 2026

Digital Observatory of the Amazon: Data Science Applied to the Analysis of Social, Economic, Environmental, and Health Indicators

October 22, 2026

Continuous Learning as a Factor of Organizational Growth in the Perception of Collaborators

October 22, 2026

Statistical Regression in Microgrids for Operational Performance Estimation: Campusgrid Case

October 22, 2026

Forecast of Aggregate Default in the Brazilian Financial System by Sarimax Model with Macroeconomic Variables

October 22, 2026

Stakeholder Engagement Factors in International Technology Transfer to a Public Hospital

A gestão financeira pessoal representa um desafio crescente, influenciada por vieses cognitivos que afetam a tomada de decisão, mesmo em indivíduos com alto conhecimento. O presente estudo analisou a relação entre o nível de educação financeira e a ocorrência de vieses comportamentais na tomada de decisão financeira de adultos residentes no Brasil. Realizou-se um levantamento de campo de natureza aplicada, quantitativa e descritivo-correlacional, por meio da aplicação de um questionário estruturado a 163 respondentes. O instrumento contemplou variáveis sociodemográficas, conhecimento e hábitos financeiros, e situações hipotéticas para identificar vieses cognitivos. Os resultados indicaram que níveis mais elevados de educação financeira associaram-se a melhores práticas de controle financeiro, poupança e investimento, com correlações significativas com renda e escolaridade. Contudo, a presença de vieses comportamentais (ancoragem, aversão à perda, escassez, otimismo/negatividade, viés do presente, status quo, excesso de confiança e contabilidade mental) persistiu mesmo entre indivíduos com maior escolaridade e renda, sem associação estatisticamente significativa com as variáveis sociodemográficas analisadas. Concluiu-se que a educação financeira, embora essencial para a organização e o controle financeiro, não foi suficiente para neutralizar a influência de vieses cognitivos. Sugere-se a incorporação de abordagens comportamentais aos programas de educação financeira.

Palavras-chave: Comportamento financeiro; Economia comportamental; Finanças pessoais; Heurísticas cognitivas; Vieses cognitivos.

October 22, 2026

Challenges and Possibilities of the First School Year of Children with ASD for Teaching Work

The inclusion of children with Autism Spectrum Disorder (ASD) in Early Childhood Education presents significant challenges, given the difficulty of adapting to new environments and the intensity of stimuli. The objective was to analyze the teaching performance in Early Childhood Education in the initial schooling process of children with ASD. A case study was conducted, using narrative interview as a data production technique. The results revealed that the teaching performance was characterized by mediation and affection practices, with the flexibility of routine and the use of playful resources as essential strategies for the child’s development and sensory organization. The importance of the school support professional for the success of the pedagogical work was also highlighted. The child’s integral development was promoted through intentional teaching actions that prioritized environmental adaptation, strengthening of bonds, and attention to the specific needs of ASD. As a contribution, a guide of pedagogical strategies aimed at welcoming children with ASD in public daycare centers was developed.

Keywords: Early Childhood Education; School Inclusion; Teacher Narratives; Inclusive Pedagogical Practices; Autism Spectrum Disorder.

October 22, 2026

Consumer Journey in the Decision to Purchase Infoproducts in the Digital Environment

The expansion of e-commerce and technological transformation have driven the need to understand consumer behavior regarding infoproducts, whose perceived experience has become crucial for professional development. The study aimed to analyze the factors that influenced the purchase decision of infoproducts in the digital environment, considering the stages of the consumer journey. An applied, quantitative, and descriptive research was conducted through a field survey with 100 consumers from Serra Gaúcha. Data collection occurred via a structured questionnaire, using non-probabilistic sampling. The results revealed a qualified audience with strong “mobile-first” behavior. It was found that the availability of detailed information and social approval were the main attraction triggers in the pre-purchase phase. In the transactional and subsequent stages, the minimization of operational friction, the agility of immediate access, and system security proved crucial for conversion and satisfaction, although gaps in human support efficiency were identified. Retention was based on the trust generated by positive past experiences, overcoming the price factor. It was concluded that the infoproduct consumer journey operates as an integrated system, requiring alignment between technological infrastructure, content curation, and support humanization to ensure engagement.

Keywords: E-commerce; Consumer behavior; Consumer experience; Loyalty; Virtual business.

October 22, 2026

Psychosocial Risks and Mental Health at Work in a Banking Institution

Mental health in the workplace is a topic of growing relevance, impacting professional development and employee well-being. A study was conducted to identify aspects and situations that can influence absenteeism levels in a banking institution. The research involved 108 participants and used a structured questionnaire, based on the short version of COPSOQ II by Gonçalves (2021), which contained 45 closed Likert scale questions and two open-ended questions. Data were collected via an online platform and analyzed with a descriptive and quantitative approach. The results revealed a scenario that indicated the need for adjustments in the distribution of activities and in managers’ perception of behaviors and signs of disorders. Important evidence was identified in the participants’ open-ended responses. It was concluded that the study contributed to promoting the perception of a safe environment for mental health in the organization.

Keywords: Absences due to disorders; Professional development; Operational leadership; Medical leave; Mental health.

October 22, 2026

Digital Observatory of the Amazon: Data Science Applied to the Analysis of Social, Economic, Environmental, and Health Indicators

The state of Amazonas faced historical challenges in consolidating the Unified Health System (SUS), stemming from geographical barriers, socioeconomic inequalities, and structural limitations. This study described the development and implementation of a health information geoportal, designed to support monitoring, evaluation, and planning in the Amazonian context. A descriptive case study approach was adopted, based on documentary analysis and secondary data, covering the mapping of functionalities, technical infrastructure, and validation steps of the tool. The Portal was developed with open technologies and integrated multiple national databases, offering interactive dashboards, thematic maps, and a digital document repository. Among the results, highlighted were the georeferenced visualization of key indicators, the possibility of temporal comparative analyses, and the use of the platform in reviewing health plans and redirecting strategic actions. The portal can represent an advance in digital governance and evidence-based management, with potential for replication in other highly complex territories, provided it is accompanied by investments in technological infrastructure, professional training, and inter-institutional coordination.

Keywords: Amazon; Health Evaluation; Digital Health; Health Information Systems.

October 22, 2026

Continuous Learning as a Factor of Organizational Growth in the Perception of Collaborators

The impact of continuous learning and knowledge management on organizational growth was studied in a context of globalization and constant changes. The study’s objective was to analyze employees’ perception of learning culture and its importance for personal and professional development. Data were obtained through an anonymous survey, with a quantitative approach, using a form with “survey” type questions applied to 87 professionals from the technology, health, education, commerce, and services sectors. The results indicated that employees had a positive perception of the contribution of continuous learning to personal and organizational growth, recognizing its benefits for the development of new skills, innovation, and competitiveness. However, it was observed that the incentive for internal knowledge exchange and the offering of practical training still presented moderate levels of agreement. It was concluded that continuous learning significantly contributes to individual and organizational development, suggesting that companies should develop strategic policies for learning focused on new skills, aiming for greater competitive advantage and innovation.

Keywords: Continuous learning; Self-development; Organizational growth; Knowledge Management; Innovation.

October 22, 2026

Statistical Regression in Microgrids for Operational Performance Estimation: Campusgrid Case

A multiple statistical regression model was proposed to estimate the operational performance of UNICAMP’s CampusGrid microgrid, based on internal electrical quantities. The microgrid integrates photovoltaic generation, batteries, and a gas generator. Energy management system data were collected at five-minute intervals over one month. The Operational Reliability Index, defined as the ratio between supplied and demanded energy, was established as the dependent variable. The model was adjusted for the islanded operation mode, evaluating the impact of demand, state of charge, solar generation, and supplied energy. The results revealed high explanatory power, indicating that operational reliability depends on the interaction between supply and demand. It was observed that the nighttime period concentrated the highest rate of partial supply, attributed to the reduction in solar generation. It was concluded that multiple regression offers an interpretable and useful framework for microgrid monitoring and control strategy support.

Keywords: Battery storage; Photovoltaic generation; Reliability Index; Islanded operation; Energy Management System.

October 22, 2026

Forecast of Aggregate Default in the Brazilian Financial System by Sarimax Model with Macroeconomic Variables

Aggregate non-payment in the Brazilian financial system is a crucial indicator of credit risk, the prediction of which is regulated by CMN Resolutions n.° 4.557/2017 and 4.966/2021 for the prospective measurement of expected losses. Aggregate non-payment was modeled and predicted using the SARIMAX model, projecting its trajectory until December 2026 under optimistic, base, and pessimistic scenarios. 164 monthly observations (July/2012 to February/2026) and macroeconomic variables such as unemployment, Selic, IPCA, IBC-Br, and household debt were used. The specification with contemporaneous variables was selected based on information criteria and coefficient stability, and residual diagnostics confirmed statistical adequacy, with a MAPE of 3.12%. Historical non-payment ranged between 4.33% and 7.64%. The estimation identified unemployment (β = 0.2534; p < 0.001) and Selic (β = 0.2128; p < 0.001) as the main positive determinants, with strong temporal persistence (AR(1) = 0.8965). Selic showed increasing correlation with non-payment (r = 0.62 at lag 12), while unemployment demonstrated a contemporaneous effect (r = 0.33). Projections indicated non-payment between 6.30% and 7.98% in December 2026, with a range of 1.68 percentage points, highlighting direct implications for provisioning in financial institutions.

Keywords: Unemployment; Predictive modeling; Monetary policy; Loss provisioning; Time series.

October 22, 2026

Stakeholder Engagement Factors in International Technology Transfer to a Public Hospital

This study described the critical factors perceived for stakeholder engagement in international technology transfer in a public hospital, with a focus on planning. The objective was to describe these conditioning factors, according to the perception of professionals. A single case study was conducted, applying a questionnaire to 25 professionals and using content analysis, complemented by descriptive tabulations. The main obstacles identified were resistance to change (72%), funding (64%), lack of communication (60%), and bureaucracy (40%). Only 28% of respondents recognized the existence of adequate tools for stakeholder management, while 48% denied it and 24% did not know. Exploratory cross-tabulations indicated variations in the assessment of influence according to time of practice, but did not establish causal relationships. Lack of engagement was associated with delays, low adherence, waste of resources, communication failures, compromised knowledge absorption, and risks to patient safety. The conclusions were limited to the perceptions of the researched group and supported preliminary recommendations for intersectoral communication and coordination.

Keywords: Case study; Project Management; Health innovation; Stakeholders; Public sector.

Economy

October 19, 2026

End of the 6×1 scale: what do the numbers allow us to conclude?

Following the promulgation of the end of the 6×1 shift system, the debate on the shorter workday moves to implementation and requires distinguishing social benefits from economic costs. The end of the 6×1 shift system transitioned from legislative discussion to a constitutional change in Brazil. On October 16, 2026, the National Congress promulgated Constitutional Amendment No. 141/2026, originating from Proposal […]

Artificial Intelligence

October 16, 2026

Why study if AI exists?

When answers become easy, learning comes to mean more than just arriving at them. Why do I need to learn this if I can ask an artificial intelligence? This type of question has accompanied education for decades. Why learn an operation if a calculator exists? Why memorize information if it can be found […]

Tax Management

October 15, 2026

Pink tax and the gender of price

How different marketing treatments for men and women are incorporated by consumption taxation Imagine a shelf on which two razors of the same brand and intended for the same purpose are displayed side by side: one has dark packaging and promises strength, precision, and efficiency; the other comes in shades of pink or […]

Health

October 13, 2026

The missing link in health transformation

Why is innovation not enough to drive changes in organizations and health systems Digital transformation has become a strategic priority for health systems worldwide. Electronic health records, artificial intelligence, telemedicine, remote monitoring, and data analysis are now part of the strategic agenda of governments and health organizations. Paradoxically, […]

Marketing

October 09, 2026

Bitcoin or pets: what Artificial Intelligence would not see

A real-life consumer intelligence case shows what gets left out of the brief A few years ago, an American insurance company hired a specialized consumer intelligence firm to answer three questions: The insurer was also considering launching paid insurance in bitcoin. At the time, the consultancy’s workflow was as follows: the analyst created a query […]

School Management

October 06, 2026

The school is not a casino: when trading cards become currency in the playground

Why, for certain students, does school manage to offer so few experiences of achievement? The World Cup is over. Brazil left the competition, the school holidays came, the final was played and the football calendar moved on. It would be reasonable to imagine that, with the end of the tournament and the return to classes, […] would also begin

Energy Transition

September 30, 2026

Renewable energies: the Brazilian advantage

Brazil needs to invest in the capacity to transform natural resources into innovation, productivity, and development. Talking about energy is talking about development. Behind industry, transportation, services, cities, and increasingly artificial intelligence, there is an energy infrastructure that sustains the economy. Understanding a country’s energy matrix is understanding […]

Data Analysis

September 30, 2026

When the data changes, has reality changed?

An analysis of almost 20 million teaching positions-year between 2020 and 2025 shows how changes in population composition and record production can alter the interpretation of historical series Between 2020 and 2025, the number of formal teaching employment positions identified in RAIS (Annual Relation of Social Information), from the Ministry of Labor, […]

Artificial Intelligence

September 28, 2026

Artificial intelligence and the trends

Why do so many people join in on the fun with ChatGPT? I was never the trend person. I still am not. Some pass through my timeline, I see them, I find them interesting but not to the point of trying them out. This was the case, inclusive, with some recent waves of images made by artificial intelligence, like the 80s trend, the one that […]

Tax Management

September 01, 2026

Tax reform and estate planning in the countryside

Efficiency in managing taxes facilitates governance and allows for the organization of family succession. Rural wealth organization has become a strategic measure in light of the tax reform in Brazil, which came into effect this year. The planning of activities and rural wealth management, through tools such as the transition from an Individual to […]

Renata de Gaspari Valdejão Almeida

Artificial Intelligence

August 05, 2026

Inside the company and off the radar: the risks of Shadow AI

According to Microsoft, 75% of professionals already use artificial intelligence at work without company approval. The use of generative artificial intelligence is already part of the corporate routine. Tools like ChatGPT, Copilot, and other platforms help professionals produce texts, analyze data, create presentations, and accelerate tasks in practically all sectors. But this advance […]

Renata de Gaspari Valdejão Almeida

Strategy

July 02, 2026

The future as a competence

Academic interest reinforces the projection of scenarios as an essential tool for decision-making. While some companies are still trying to get the next step right, others are already working with multiple possible futures — and preparing for all of them. But after all, what does it mean to “think about the future”? Is it a trend, a strategy, or a new competence that […]

Renata de Gaspari Valdejão Almeida

Innovation

Technology

May 21, 2026

Agro SP Corridor: innovation, institutional articulation and future

Pecege commands the executive operation of the 400 km ecosystem activities that became known as the Brazilian agro Silicon Valley In 2025, in addition to taking over the management of the Piracicaba Technology Park, the Pecege Institute was also designated to command the executive operation of the State Agribusiness Innovation Corridor activities, the so-called […]

Education

April 27, 2026

How to turn a TCC into a business

New management of the Piracicaba Technology Park offers consulting to MBA graduates The new project director of the Piracicaba Technology Park (PTP), Alexandre Barreto, often says that one of the greatest rewards he has ever had in life was seeing his MBA USP/Esalq students structure companies from their final projects […]

Technology

March 20, 2026

The human repertoire in the age of AI

Critical thinking and creativity are irreplaceable in the work of a marketing professional Artificial Intelligence (AI) is capable of incredible feats, but it does not yet possess the judgment, empathy, intuition, and strategic creativity necessary to do the job of a good marketing professional. In this area, it is important to know how to use the tool to automate […]

Renata de Gaspari Valdejão Almeida

Agribusiness

Innovation

February 23, 2026

ESG as a growth strategy in a cooperative in RS

Santa Clara balances social and environmental responsibility with innovation Founded in 1911 by Italian immigrants in Carlos Barbosa, in the Serra Gaúcha region, Cooperativa Santa Clara is an example that it is possible to balance economic growth with social and environmental responsibility. From a small cheese production company, the organization has evolved into a […]

Renata de Gaspari Valdejão Almeida

Agribusiness

Innovation

January 22, 2026

Brazil free of Foot-and-Mouth Disease: what changes for Brazilian beef?

With a herd of over 200 million head of cattle, the country is currently the leader in exports. Brazil recently received, for the first time, the certification of country free of foot -and-mouth disease without vaccination, conferred by the World Organisation for Animal Health (WOAH). The seal strengthens the country’s position as a reliable supplier of meat for the […]

Renata de Gaspari Valdejão Almeida

Innovation

December 09, 2025

Innovation: the new cycle of the Piracicaba Technology Park

Instituto Pecege assumes management, betting on the creation of new businesses Starting in 2026, Instituto Pecege will assume the management of the Piracicaba Technology Park (PTP). Among its responsibilities will be the administration, operation, maintenance, and expansion of the park, in addition to fostering activities in innovation, entrepreneurship, and science and technology. The choice was made […]

Renata de Gaspari Valdejão Almeida

July 06, 2026

Large-Scale Architecture: DSR-SCRUM Change Management

Mônica Alessandra Guerios; Maria do Carmo Assis Todorov Abstract generated by ResumeAI, an artificial intelligence solution developed by Instituto Pecege focused on synthesis and writing. Architecture and urbanism are configured as essential bases in an increasingly complex and challenging urban space, where the intrinsic connection between social, environmental, and technological issues […]