Utilizando Algoritmos de Deep Learning para a Previsão da Taxa de Inflação

Authors

DOI:

https://doi.org/10.20397/2177-6652/2025.v25i3.3182

Keywords:

Inflação, Previsões, Machine Learning

Abstract

Abstract

Objective: This article demonstrates the predictions of machine learning models such as RNNs and LSTMs, being applied to predict the inflation rate.

Originality/Relevance: The use of machine learning models to predict price changes is widely discussed in the industry. This article contributes to the advancement of the area by analyzing parameters and their results individually and jointly, showing new results and serving as a guide for future projects.

Methodology/Approach: Data from the United States were used, with the parameters being the CPI, the federal funds interest rate, the growth of Real GDP and the velocity of M2. Where the database was normalized and divided into 3 smaller batches, and in each batch predictions were made, and these predictions were compared with the test values.

Main Results: All parameters used were shown to be efficient in assisting in forecasts beyond the database, generating forecasts with correlations above 0.65% and MAE errors lower than 3.

Theoretical/methodological contributions: Demonstration of the weight of each parameter for the results, and economic explanations about changes in the parameters and how these changes affect the predictive capabilities of the models.

Social Contributions: This study contributes both to industries that need good forecasts about future inflation, and researchers who want to create more accurate models.

 

Keywords: Inflation, Forecasts, Machine Learning.

Author Biographies

Hygor Santiago Lara, Universidade Estatual de Campinas (UNICAMP)

Doutorando em Engenharia Mecânica pela Universidade de Campinas (UNICAMP), trabalha com Inteligência Artificial e projetos mecânicos. Professor (PED) no curso de graduação em Engenharia Mecânica na UNICAMP, curso de Vibrações Mecânicas sob orientação do Prof. Alberto Serpa. Professor orientador e fundador do Grupo de Estudos em Inteligência Artificial (GEIA) desenvolvendo trabalhos inter disciplinares com alunos de graduação e graduados. Professor e coordenador didático dos cursos de Programação, Dados e Inteligência Artificial do IESAE. Membro do comitê técnico e científico da revista Gestão e Tecnologia. Membro de projetos de pesquisa em parceria: a) Bancada de vibrações para diagnostico de defeitos mecânicos 4.0 (parceria com LASID-UFSJ), b) Simulação de veículos autônomos (parceria Milhas Gerais - UFSJ), leitura de partituras e produção musical assistida por IA (parceria Orquestra do Conservatório Pe. José Maria Xavier). Mestrado com aplicação de Inteligência Artificial no diagnostico de doenças cardíacas por eletrocardiografia. Pós-graduado em Ciências de Dados pela Digital House e Inteligência Artificial pela Data Science Academy, com experiência no núcleo de tecnologia das empresas Thinkseg e Bidu onde atuou por 2 anos. Participa do Laboratório de Dinâmica de Estruturas e Máquinas sob orientação do Professor Milton Dias Júnior. Graduado em Engenharia Mecânica pela Universidade Federal de São João Del Rei (UFSJ). (Texto informado pelo autor)

José Eduardo Mesquita Junior, Sou graduado em Engenharia Mecânica pela UNICAMP, e tenho Mestrado Profissional em Administração pela FPL-Fundação Pedro Leopoldo

Graduação em Engenharia Mecânica pela Universidade Estadual de Campinas-UNICAMP (1996). Mestrado Profissional em Administração de Empresas pela Fundação Pedro Leopoldo (2022). Especialização em Economia de Empresas (USP), Gestão Econômica e Financeira (FIPE) e Docência do Ensino Superior (Unopar). Atualmente ocupa a posição de Coordenador de Engenharia de Suporte ao Cliente - IDV Latin America (IVECO Group). Experiência profissional na área de Engenharia Mecânica, atuando principalmente nos seguintes temas: Engenharia Mecânica, Engenharia de Manutenção, Engenharia Aeronáutica, Custos de Manutenção e Suporte Técnico. (Texto informado pelo autor)

References

Ahmad, A.M; Ismail, S.; Samaon, D.F. (2004). Recurrent neural network with backpropagation through time for speech recognition. IEEE International Symposium on Communications & Information Technology, 1, 98-102.

Bach G. L.; Albert A. (1957). The Redistributional Effects of Inflation. The Review of Economics and Statistics Vol. 39, No. 1 (Feb, 1957), pp. 1-13 (13 pages).

Bagus, Philipp. (2009). The Quality of Money The Quartely Journal of Austrian Economics 12, No. 4 (2009): 22–45.

Chandak A.; Prajwal G.; Shyam S.; Ayushi B.; Soumya T. (2019). Car Price Prediction Using Machine Learning. International Journal of Computer Sciences and Engineering.

Hanke, S. H.; Alex K. F. Kwok. (2009) On the Measurement of Zimbabwe's Hyperinflation. Cato Journal Vol. 29, No.2 (Spring/Summer 2009).

Hao X.; Guigang Z; Shang M. (2016). Deep Learning. International Journal Of Semantic Computing Vol. 10, No. 03, pp. 417-439(2016).

Knight, Alex. (2002) A Bubble on the Mighty Mississippi: An Application of a General Model of Speculative Bubbles to the Mississippi Bubble of 1716-1720. Major Themes in Economics, 4, 71-94.

Koutnik .J; Klaus G.; Faustino G.; Juergen S. (2014). A Clockwork RNN. 31st International Conference on Machine Learning, PMLR 32(2):1863-1871.

Pal S.; Soumadip G.; Amitava N. (2018). Sentiment Analysis in the Light of LSTM Recurrent Neural Networks. International Journal of Synthetic Emotions, Volume 9, Issue.

Sakib-Bin-Amin. (2011). Quantity Theory of Money and its Applicability: The Case of Bangladesh. World Review of Business Research Vol. 1. No. 4. September 2011. Pp.33-4.

Downloads

Published

2025-06-30

How to Cite

Oliveira Santos, G., Santiago Lara, H., & Mesquita Junior, J. E. (2025). Utilizando Algoritmos de Deep Learning para a Previsão da Taxa de Inflação. Revista Gestão & Tecnologia, 25(3), 314–326. https://doi.org/10.20397/2177-6652/2025.v25i3.3182

Issue

Vol. 25 No. 3 (2025)

Section

ARTIGO

License

Copyright (c) 2025 Revista Gestão & Tecnologia

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Authors who publish with this journal agree to the following terms: • 1. The author(s) authorize the publication of the article in the journal. • 2. The author(s) ensure that the contribution is original and unpublished and is not being evaluated in other journal(s). • 3. The journal is not responsible for the opinions, ideas and concepts expressed in the texts because they are the sole responsibility of the author(s). • 4. The publishers reserve the right to make adjustments and textual adaptation to the norms of APA. • 5. Authors retain copyright and grant the journal right of first publication, with the work [SPECIFY PERIOD OF TIME] after publication simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal. • 6. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal. • 7. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access) at http://opcit.eprints.org/oacitation-biblio.html