Journal of Korean Society for Quality Management (품질경영학회지)

Korean Society for Quality Management (한국품질경영학회)
권호 표지
DOI QR코드

DOI QR Code

Forecasting Baltic Dry Index by Implementing Time-Series Decomposition and Data Augmentation Techniques

시계열 분해 및 데이터 증강 기법 활용 건화물운임지수 예측

  • Han, Min Soo (Department of Shipping Management, Graduate School of Korea Maritime and Ocean University) ;
  • Yu, Song Jin (Department of Shipping Management and Economics, Korea Maritime and Ocean University)
  • 한민수 (한국해양대학교 일반대학원 해운경영학과 ) ;
  • 유성진 (한국해양대학교 해양인문사회과학대학 해양경영경제학부)
  • Received : 2022.09.06
  • Accepted : 2022.11.07
  • Published : 2022.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Purpose: This study aims to predict the dry cargo transportation market economy. The subject of this study is the BDI (Baltic Dry Index) time-series, an index representing the dry cargo transport market. Methods: In order to increase the accuracy of the BDI time-series, we have pre-processed the original time-series via time-series decomposition and data augmentation techniques and have used them for ANN learning. The ANN algorithms used are Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) to compare and analyze the case of learning and predicting by applying time-series decomposition and data augmentation techniques. The forecast period aims to make short-term predictions at the time of t+1. The period to be studied is from '22. 01. 07 to '22. 08. 26. Results: Only for the case of the MAPE (Mean Absolute Percentage Error) indicator, all ANN models used in the research has resulted in higher accuracy (1.422% on average) in multivariate prediction. Although it is not a remarkable improvement in prediction accuracy compared to uni-variate prediction results, it can be said that the improvement in ANN prediction performance has been achieved by utilizing time-series decomposition and data augmentation techniques that were significant and targeted throughout this study. Conclusion: Nevertheless, due to the nature of ANN, additional performance improvements can be expected according to the adjustment of the hyper-parameter. Therefore, it is necessary to try various applications of multiple learning algorithms and ANN optimization techniques. Such an approach would help solve problems with a small number of available data, such as the rapidly changing business environment or the current shipping market.

Keywords

References

  1. Abdollahi, H. 2020. A novel hybrid model for forecasting crude oil price based on time series decomposition. Applied energy, 267, 115035.
  2. Allaire, JJ. & Chollet, F. 2022. Package 'keras'. R Interface to 'Keras'.
  3. Allaire, JJ. & Tang, Y. 2022. Package 'tensorflow'. R Interface to 'TensorFlow'.
  4. Benkachcha, S., Benhra, J., & El Hassani, H. 2015. Seasonal time series forecasting models based on artificial neural network. International Journal of Computer Applications 116(20).
  5. Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. 1990. STL: A seasonal-trend decomposition. J. Off. Stat 6(1):3-73.
  6. Demir, S., Mincev, K., Kok, K., & Paterakis, N. G. 2021. Data augmentation for time series regression: Applying transformations, autoencoders and adversarial networks to electricity price forecasting. Applied Energy 304: 117695.
  7. Fritsch, S., Guenther, F., & Guenther, M. F. 2019. Package 'neuralnet'. Training of Neural Networks.
  8. Gers, F. A., Eck, D., & Schmidhuber, J. 2002. Applyting LSTM to time series predictable through time-window approaches. In Neural Nets WIRN Vietri-01, pp. 193-200. Springer, London.
  9. Hafen, R. 2016. Package 'stlplus'. Ehanced Seasonal Decomposition of Time Series by Loess.
  10. HAN, M. & Yu, S. J. 2019. Prediction of Baltic Dry Index by Applications of Long Short-Term Memory. Journal of the Korean Society for Quality Management 47(3):497-508.
  11. Hansen, J. V. & Nelson, R. D. 2003. Forecasting and recombining time-series components by using neural networks. Journal of the Operational Research Society 54(3):307-317. https://doi.org/10.1057/palgrave.jors.2601523
  12. Iwana, B. K. & Uchida, S. 2021. An empirical survey of data augmentation for time series classification with neural networks. Plos one 16(7):e0254841.
  13. Kang, S., Cho, K., & Na, M. 2021. Forecasting Crop Yield Using Encoder-Decoder Model with Attention. Journal of the Korean Society for Quality Management 49(4):569-579.
  14. Khandelwal, I., Adhikari, R., & Verma, G. 2015. Time series forecasting using hybrid ARIMA and ANN models based on DWT decomposition. Procedia Computer Science 48:173-179. https://doi.org/10.1016/j.procs.2015.04.167
  15. Lee, S. W. & Kim, H. Y. 2020. Stock market forecasting with super-high dimensional time-series data using ConvLSTM, trend sampling, and specialized data augmentation. Expert Systems with Applications 161:113704.
  16. Lin, Y., Koprinska, I., & Rana, M. 2021. SSDNet: State space decomposition neural network for time series forecasting. In 2021 IEEE International Conference on Data Mining (ICDM), pp. 370-378. IEEE.
  17. Mendez-Jimenez, I., & Cardenas-Montes, M. 2018.. Time series decomposition for improving the forecasting performance of convolutional neural networks. In Conference of the Spanish Association for Artificial Intelligence (pp. 87-97). Springer, Cham.
  18. Oh, C., Han, S. & Jeong, J. 2020. Time-series data augmentation based on interpolation. Procedia Computer Science 175:64-71. https://doi.org/10.1016/j.procs.2020.07.012
  19. Oliveira, D. D., Rampinelli, M., Tozatoo, G. Z., Andreao, R. V., & Muller, S. M. 2021. Forecasting vehicular traffic flow using MLP and LSTM. Neural Computing and applications 33(24):17245-17256. https://doi.org/10.1007/s00521-021-06315-w
  20. Ouyang, Z., Ravier, P., & Jabloun, M. 2021. STL Decomposition of Time Series Can Benefit Forecasting Done by Statistical Methods but Not by Machine Learning Ones. Engineering Proceedings 5(1):42.
  21. Quast, B. & Fichou, D. (2022). Package 'rnn'. Recurrent Nerual Network.
  22. Wen, Q., Sun, L., Yang, F., Song, X., Gao, J., Wang, X., & Xu, H. 2020. Time series data augmentation for deep learning: A survey. arXiv preprint arXiv:2002.12478.
  23. Zhang, G. P. 2003. Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing 50: 159-175. https://doi.org/10.1016/S0925-2312(01)00702-0