Prediction of exports by department in Peru using Machine Learning models and time series
Article Sidebar
Main Article Content
Abstract
This study presented an application of machine learning models for predicting exports by department in Peru, using annual data from the Central Reserve Bank of Peru (BCRP) between 2005 and 2022. The CatBoost, XGBoost, Linear Regression, and Prophet algorithms were evaluated, comparing their performance using the MAE, RMSE, MAPE, and R² metrics. The results showed that the CatBoost model performed superiorly, with an R² of 96.28% and an MAE of 32.91 in validation. Prophet followed, achieving a MAPE of 3.75% and an RMSE of 72.15, demonstrating high accuracy in time series. In contrast, Linear Regression presented an R² of 82.72%, reflecting limitations in modeling nonlinear relationships. The XGBoost model, while competitive, showed an R² of 91.74% with an RMSE of 163.58. Additionally, a functional web prototype was developed using Django and React for visualizing results, which enabled the generation of dynamic predictions by product and department. Taken together, the findings demonstrated the effectiveness of machine learning for regional economic analysis and the formulation of evidence-based export policies.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Los nombres y las direcciones de correo electrónico introducidos en esta revista se usarán exclusivamente para los fines establecidos en ella y no se proporcionarán a terceros o para su uso con otros fines.
How to Cite
References
P. Suler, Z. Rowland, y T. Krulicky, «Evaluation of the Accuracy of Machine Learning Predictions of the Czech Republic’s Ex-ports to the China», Journal of Risk and Financial Management, vol. 14, n.o 2, feb. 2021, doi: 10.3390/jrfm14020076.
H. Jošić y B. Žmuk, «A Machine Learning Approach to Forecast International Trade: The Case of Croatia», Business Systems Research, vol. 13, n.o 3, pp. 144-160, oct. 2022, doi: 10.2478/bsrj-2022-0030.
M. Iaousse, Y. Jouilil, M. Bouincha, y D. Mentagui, «A Comparative Simulation Study of Classical and Machine Learning Tech-niques for Forecasting Time Series Data», International journal of online and biomedical engineering, vol. 19, n.o 8, pp. 56-65, 2023, doi: 10.3991/ijoe.v19i08.39853.
Y. Gulzar et al., «Predicting High Technology Exports of Countries for Sustainable Economic Growth by Using Machine Learn-ing Techniques: The Case of Turkey», Sustainability 2024, Vol. 16, Page 5601, vol. 16, n.o 13, p. 5601, jun. 2024, doi: 10.3390/SU16135601.
B. Sellami, C. Ounoughi, T. Kalvet, M. Tiits, y D. Rincon-Yanez, «Harnessing Graph Neural Networks to Predict International Trade Flows», Big Data and Cognitive Computing, vol. 8, n.o 6, jun. 2024, doi: 10.3390/bdcc8060065.
S. K. R, B. M.S., y V. Pachava, «Machine Learning-Driven Export Forecasting: A Comparative Analysis for MSME Growth», Journal of Economics, Innovative Management and Entrepreneurship, vol. 3, n.o 1, feb. 2025, doi: 10.59652/JEIME.V3I1.401.
C. Maier, J. B. Thatcher, V. Grover, y Y. K. Dwivedi, «Cross-sectional research: A critical perspective, use cases, and recom-mendations for IS research», Int J Inf Manage, vol. 70, p. 102625, jun. 2023, doi: 10.1016/J.IJINFOMGT.2023.102625.
M. González Mares, «Hernández-Sampieri, R. & Mendoza, C (2018). Metodología de la investigación. Las rutas cuantitativa, cualitativa y mixta», Revista Universitaria Digital de Ciencias Sociales (RUDICS), vol. 10, n.o 18, pp. 92-95, ene. 2019, doi: 10.22201/FESC.20072236E.2019.10.18.6.
«Exportaciones e importaciones». Accedido: 25 de octubre de 2025. [En línea]. Disponible en: https://estadisticas.bcrp.gob.pe/estadisticas/series/anuales/exportaciones-e-importaciones
F. H. Garabaghi, S. Benzer, y R. Benzer, «Modeling dissolved oxygen concentration using machine learning techniques with dimensionality reduction approach», Environ Monit Assess, vol. 195, n.o 7, pp. 1-23, jul. 2023, doi: 10.1007/S10661-023-11492-3/METRICS.
J. K. Sayyad, K. Attarde, y N. Saadouli, «Optimizing e-Commerce Supply Chains With Categorical Boosting: A Predictive Mod-eling Framework», IEEE Access, vol. 12, pp. 134549-134567, 2024, doi: 10.1109/ACCESS.2024.3447756.
«CatBoost in Machine Learning - GeeksforGeeks». Accedido: 25 de octubre de 2025. [En línea]. Disponible en: https://www.geeksforgeeks.org/machine-learning/catboost-ml/
T. Zhao, G. Chen, S. Suraphee, T. Phoophiwfa, y P. Busababodhin, «A hybrid TCN-XGBoost model for agricultural product market price forecasting», PLoS One, vol. 20, n.o 5, p. e0322496, may 2025, doi: 10.1371/JOURNAL.PONE.0322496.
«Cómo funciona el XGBoost algoritmo de SageMaker IA - Amazon SageMaker AI». Accedido: 25 de octubre de 2025. [En línea]. Disponible en: https://docs.aws.amazon.com/es_es/sagemaker/latest/dg/xgboost-HowItWorks.html
E. Aytaç, «Forecasting Turkey’s Hazelnut Export Quantities with Facebook’s Prophet Algorithm and Box-Cox Transformation», ADCAIJ: Advances in Distributed Computing and Artificial Intelligence Journal, vol. 10, n.o 1, pp. 33-47, feb. 2021, doi: 10.14201/ADCAIJ20211013347.
C. Guo, Q. Ge, H. Jiang, G. Yao, y Q. Hua, «Maximum Power Demand Prediction Using Fbprophet with Adaptive Kalman Filtering», IEEE Access, vol. 8, pp. 19236-19247, 2020, doi: 10.1109/ACCESS.2020.2968101.
A. B. Fawait, S. Rahmah, A. Diana, S. Da Costa, N. Insyroh, y A. A. Firdaus, «Implementation of Data Mining Using Simple Linear Regression Algorithm to Predict Export Values», Scientific Journal of Engineering Research, vol. 1, n.o 1, pp. 26-32, ene. 2025, doi: 10.64539/SJER.V1I1.2025.11.
D. H. Min y H. K. Yoon, «Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping», Sci Rep, vol. 11, n.o 1, dic. 2021, doi: 10.1038/S41598-021-86137-X.