American Journal of Neural Networks and Applications

Research Article | | Peer-Reviewed |

Modelling and Forecasting Inflation Rates in Kenya Using ARIMA-ANN Hybrid Model

Received: Sep. 27, 2023    Accepted: Oct. 16, 2023    Published: Oct. 28, 2023
Views:       Downloads:

Share

Abstract

This study explored the complexities of modeling and forecasting inflation rates in Kenya, leveraging a sophisticated ARIMA-ANN hybrid model. Traditional ARIMA models, although proficient in capturing linear relationships, often falter in the face of non-linear, complex patterns inherent in economic data. To enhance accuracy, we integrated an ANN with a specifically chosen ARIMA (1, 0, 11) model, benefiting from ANN’s capability to delineate non-linear correlations and intricacies. This hybrid model was meticulously trained to minimize the MSE, epitomizing efficiency in both training and validation phases. Empirical results showcased the model’s commendable predictive accuracy. A comparative analysis accentuated its supremacy over the traditional ARIMA model, delineated by superior MSE, RMSE, MAE, and MAPE metrics. The hybrid model adeptly amalgamated ARIMA’s statistical robustness with ANN’s adeptness at non-linear pattern recognition, ensuring enhanced forecast precision. The model is not just a theoretical construct but a pragmatic tool, instrumental for policymakers, economists, and stakeholders, offering insightful foresights that are pivotal for strategic planning and decision-making. The forecasting accuracy of our hybrid model was rigorously tested against actual inflation data, and its performance metrics underscored reliability and precision. Future research could potentially augment this model by integrating more advanced neural network architectures, and incorporating external economic indicators to further enhance forecasting accuracy. This study is a substantial stride towards a nuanced understanding of inflation dynamics in Kenya, offering tools that are not only statistically robust but also practically applicable in real-world economic scenarios. This intricate blend of statistical and machine learning techniques promises to be a cornerstone for future economic forecasting endeavors.

DOI 10.11648/j.ajnna.20230901.12
Published in American Journal of Neural Networks and Applications ( Volume 9, Issue 1, June 2023 )
Page(s) 8-17
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Inflation, ARIMA-ANN, Time Series, Forecasting, Modelling, ANN

References
[1] Adebiyi, A. A., Adewumi, A. O., & Ayo, C. K. (2014). Comparison of arima and artificial neural networks models for stock price prediction. Journal of Applied Mathematics, 2014.
[2] Creswell, J. W., & Creswell, J. D. (2017). Research design: Qualitative, quantitative, and mixed methods approaches. Sage publications.
[3] Devi, K., & Monika, et. al. (2021). Forecasting of wheat production in haryana using hybrid time series model. Journal of Agriculture and Food Research, 5, 100-175.
[4] Elwasify, A. I. (2015). A combined model between artificial neural networks and arima models. International Journal of Recent Research in Commerce Economics and Management, 2 (2), 134-140.
[5] Jamil, H. (2022). Inflation forecasting using hybrid arima-lstm model (Unpublished doctoral dissertation). Laurentian University of Sudbury.
[6] Khan, F., Urooj, A., & Muhammadullah, S. (2021). An arima-ann hybrid model for monthly gold price forecasting: empirical evidence from pakistan. Pakistan Econ Rev, 4 (1), 61-75.
[7] Koutroumanidis, T., Ioannou, K., & Arabatois, G. (2009). Predicting fuelwood prices in greece with the use of arima models, artificial neural networks and a hybrid arimaannmodel. Energy Policy, 37 (9), 3627-3634.
[8] Lidiema, C. (2017). Modelling and forecasting inflation rate in kenya using sarima and holt-winters triple exponential smoothing. American Journal of Theoretical and Applied Statistics, 6 (3), 161-169.
[9] Mucaj, R., & Sinaj, V. (2017). Exchange rate forecasting using arima, nar and arima-ann hybrid model. Exchange, 4 (10), 8581-8586.
[10] Musa, Y., & Joshua, S. (2020). Analysis of arima-artificial neural network hybrid model in forecasting of stock market returns. Asian Journal of Probability and Statistics, 6 (2), 42-53.
[11] Nyoni, T. (2018). Modeling and forecasting inflation in kenya: Recent insights from arima and garch analysis. Dimorian Review, 5 (6), 16-40.
[12] Siamba, S. N. (2022). Forecasting tuberculosis infections using arima and hybrid neural network models among children below 15 years in homa bay and turkana counties, Kenya (Unpublished doctoral dissertation). University of Eldoret.
[13] Uwilingiyimana, C., Munga'tu, J., & Harerimana, J. (2015). Forecasting inflation in kenya using arima-garch models. International Journal of Management and Commerce Innovations, 3 (2), 15-27.
[14] Zhang, G. P., & Qi, M. (2005). Neural network forecasting for seasonal and trend time series. European Journal of Operational Research, 16 (2), 501-514.
[15] Çavuş Büyükşahin, & Ertekin, (2018). Improving forecasting accuracy of time series data using a new arima-ann hybrid method and empirical mode decomposition. arXiv e-prints.
Cite This Article
  • APA Style

    Barry Agingu Jagero, Thomas Mageto, Samuel Mwalili. (2023). Modelling and Forecasting Inflation Rates in Kenya Using ARIMA-ANN Hybrid Model. American Journal of Neural Networks and Applications, 9(1), 8-17. https://doi.org/10.11648/j.ajnna.20230901.12

    Copy | Download

    ACS Style

    Barry Agingu Jagero; Thomas Mageto; Samuel Mwalili. Modelling and Forecasting Inflation Rates in Kenya Using ARIMA-ANN Hybrid Model. Am. J. Neural Netw. Appl. 2023, 9(1), 8-17. doi: 10.11648/j.ajnna.20230901.12

    Copy | Download

    AMA Style

    Barry Agingu Jagero, Thomas Mageto, Samuel Mwalili. Modelling and Forecasting Inflation Rates in Kenya Using ARIMA-ANN Hybrid Model. Am J Neural Netw Appl. 2023;9(1):8-17. doi: 10.11648/j.ajnna.20230901.12

    Copy | Download

  • @article{10.11648/j.ajnna.20230901.12,
      author = {Barry Agingu Jagero and Thomas Mageto and Samuel Mwalili},
      title = {Modelling and Forecasting Inflation Rates in Kenya Using ARIMA-ANN Hybrid Model},
      journal = {American Journal of Neural Networks and Applications},
      volume = {9},
      number = {1},
      pages = {8-17},
      doi = {10.11648/j.ajnna.20230901.12},
      url = {https://doi.org/10.11648/j.ajnna.20230901.12},
      eprint = {https://download.sciencepg.com/pdf/10.11648.j.ajnna.20230901.12},
      abstract = {This study explored the complexities of modeling and forecasting inflation rates in Kenya, leveraging a sophisticated ARIMA-ANN hybrid model. Traditional ARIMA models, although proficient in capturing linear relationships, often falter in the face of non-linear, complex patterns inherent in economic data. To enhance accuracy, we integrated an ANN with a specifically chosen ARIMA (1, 0, 11) model, benefiting from ANN’s capability to delineate non-linear correlations and intricacies. This hybrid model was meticulously trained to minimize the MSE, epitomizing efficiency in both training and validation phases. Empirical results showcased the model’s commendable predictive accuracy. A comparative analysis accentuated its supremacy over the traditional ARIMA model, delineated by superior MSE, RMSE, MAE, and MAPE metrics. The hybrid model adeptly amalgamated ARIMA’s statistical robustness with ANN’s adeptness at non-linear pattern recognition, ensuring enhanced forecast precision. The model is not just a theoretical construct but a pragmatic tool, instrumental for policymakers, economists, and stakeholders, offering insightful foresights that are pivotal for strategic planning and decision-making. The forecasting accuracy of our hybrid model was rigorously tested against actual inflation data, and its performance metrics underscored reliability and precision. Future research could potentially augment this model by integrating more advanced neural network architectures, and incorporating external economic indicators to further enhance forecasting accuracy. This study is a substantial stride towards a nuanced understanding of inflation dynamics in Kenya, offering tools that are not only statistically robust but also practically applicable in real-world economic scenarios. This intricate blend of statistical and machine learning techniques promises to be a cornerstone for future economic forecasting endeavors.
    },
     year = {2023}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Modelling and Forecasting Inflation Rates in Kenya Using ARIMA-ANN Hybrid Model
    AU  - Barry Agingu Jagero
    AU  - Thomas Mageto
    AU  - Samuel Mwalili
    Y1  - 2023/10/28
    PY  - 2023
    N1  - https://doi.org/10.11648/j.ajnna.20230901.12
    DO  - 10.11648/j.ajnna.20230901.12
    T2  - American Journal of Neural Networks and Applications
    JF  - American Journal of Neural Networks and Applications
    JO  - American Journal of Neural Networks and Applications
    SP  - 8
    EP  - 17
    PB  - Science Publishing Group
    SN  - 2469-7419
    UR  - https://doi.org/10.11648/j.ajnna.20230901.12
    AB  - This study explored the complexities of modeling and forecasting inflation rates in Kenya, leveraging a sophisticated ARIMA-ANN hybrid model. Traditional ARIMA models, although proficient in capturing linear relationships, often falter in the face of non-linear, complex patterns inherent in economic data. To enhance accuracy, we integrated an ANN with a specifically chosen ARIMA (1, 0, 11) model, benefiting from ANN’s capability to delineate non-linear correlations and intricacies. This hybrid model was meticulously trained to minimize the MSE, epitomizing efficiency in both training and validation phases. Empirical results showcased the model’s commendable predictive accuracy. A comparative analysis accentuated its supremacy over the traditional ARIMA model, delineated by superior MSE, RMSE, MAE, and MAPE metrics. The hybrid model adeptly amalgamated ARIMA’s statistical robustness with ANN’s adeptness at non-linear pattern recognition, ensuring enhanced forecast precision. The model is not just a theoretical construct but a pragmatic tool, instrumental for policymakers, economists, and stakeholders, offering insightful foresights that are pivotal for strategic planning and decision-making. The forecasting accuracy of our hybrid model was rigorously tested against actual inflation data, and its performance metrics underscored reliability and precision. Future research could potentially augment this model by integrating more advanced neural network architectures, and incorporating external economic indicators to further enhance forecasting accuracy. This study is a substantial stride towards a nuanced understanding of inflation dynamics in Kenya, offering tools that are not only statistically robust but also practically applicable in real-world economic scenarios. This intricate blend of statistical and machine learning techniques promises to be a cornerstone for future economic forecasting endeavors.
    
    VL  - 9
    IS  - 1
    ER  - 

    Copy | Download

Author Information
  • Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

  • Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

  • Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

  • Section