Forecasting of Dry Freight Index Data by Using Machine Learning Algorithms

Full Text (PDF, 902KB), PP.35-43

Views: 0 Downloads: 0


Kemal Akyol 1,*

1. Department of Computer Engineering, Kastamonu University, Kastamonu, 37100, Turkey

* Corresponding author.


Received: 9 Mar. 2019 / Revised: 13 Apr. 2019 / Accepted: 9 May 2019 / Published: 8 Aug. 2019

Index Terms

Crude oil price, freight index data, machine learning, Gradient Boosted Trees, Multi-Layer Perceptron


Discovery of meaningful information from the data and design of an expert system are carried out within the frame of machine learning. Supervised learning is used commonly in practical machine learning. It includes basically two stages: a) the training data are sent to as input to the classifier algorithms, b) the performance of pre-learned algorithm is tested on the testing data. And so, knowledge discovery is carried out through the data. In this study, the analysis of Lloyd data is performed by utilizing Gradient Boosted Trees and Multi-Layer Perceptron learning algorithms. Lloyd data consist of the Baltic Dry Index, Capesize Index, Panamax Index and Supramax Index values, updated daily. Accurate prediction of these data is very important in order to eliminate the risks of commercial organization. Eight datasets from Lloyd data are obtained within the frame of two scenarios: a) the last three index values in the freight index datasets; b) the last three index values in both crude oil price and freight index datasets. The results show that the models designed with Gradient Boosted Trees and Multi-Layer Perceptron algorithms are successful for Lloyd data prediction and so proved their applicability.

Cite This Paper

Kemal Akyol, "Forecasting of Dry Freight Index Data by Using Machine Learning Algorithms", International Journal of Intelligent Systems and Applications(IJISA), Vol.11, No.8, pp.35-43, 2019. DOI:10.5815/ijisa.2019.08.04


[1]B. Şahin, S. Gürgen, B. Ünver, and İ. Altın, “Forecasting the baltic dry index by using an artificial neural network approach,” Turk J Elec Eng & Comp Sci, vol. 26, 2018, pp. 1673 – 1684.
[2]Q. Han, B. Yan, G. Ning, and B. Yu, “Forecasting dry bulk freight index with improved SVM,” Math Probl in Eng, vol. 2014, Article ID 460684, 2014. doi:
[3]F. Guan, Z. Peng, K. Wang, X. Song, and J. Gao, “Multi-step hybrid prediction model of baltic supermax index based on support vector machine,” Neural Netw World, vol. 26, 2016, pp. 219-232. doi: 0.14311/NNW.2016.26.012.
[4]H.L. Wong, “BDI forecasting based on fuzzy set theory, grey system and ARIMA,” In: Ali M., Pan JS., Chen SM., Horng MF. (eds) Modern Advances in Applied Intelligence, Lect Notes Comput Sc, vol. 8482, 2014, Springer, Cham.
[5]J.H. Friedman, “Greedy Function Approximation: A Gradient Boosting Machine,” The Annals of Statistics, vol. 29, 2001, pp. 1189-1232.
[6]F. Rosenblatt, “Principles of neurodynamics: Perceptions and the theory of brain mechanism,” Spartan Books, Washington, DC, 1961.
[7]C. Ming-Tao, “A fuzzy time series model to forecast the BDI,” Fourth International Conference on Networked Computing and Advanced Information Management, vol. 2, 2008, pp. 50-53.
[8]G. Giannarakis, C. Lemonakis, A. Sormas, and C. Georganakis, “The effect of baltic dry index, gold, oil and USA,” Trade Balance on Dow Jones Sustainability Index World, vol. 7, 2017; pp. 155-160.
[9]H. Geman and W.O. Smith, “Shipping markets and freight rates: an analysis of the baltic dry index,” The Journal of Alternative Investments, vol. 5, 2012, pp. 98-109.
[10]V. Tsioumas, S. Papadimitriou, Y. Smirlis, and S. Z. Zahran, “A novel approach to forecasting the bulk freight market,” The Asian Journal of Shipping and Logistics, vol. 33, 2017, pp. 33-41. doi:
[11]Q. Ruan, Y. Wang, X. Lu, and J. Qin, “Cross-correlations between baltic dry index and crude oil prices,” Physica A, vol. 453, 2016, pp. 278-289. doi:
[12]P. Baltyn, “Baltic dry index as economic leading indicator in the United States,” Managing Innovation and Diversity in Knowledge Society Through Turbulent Time Proceedings of the Make Learn and TIIM Joint International Conference, 25-27 May 2016; Timisoara, Romania.
[13]Q. Zeng, C. Qu, A.K. Ng, and X. Zhao, “A new approach for baltic dry index forecasting based on empirical mode decomposition and neural networks,” Marit Econ Logist, vol. 18, 2016, pp. 192-210.
[14]J. Bao, L. Pan, and Y. Xie, “A new BDI forecasting model based on support vector machine,” Information Technology, Networking, Electronic and Automation Control Conference, 20-22 May 2016, Chongqing, China: IEEE.
[15]N. Apergis and J.E. Payne, “New evidence on the information and predictive content of the baltic dry index,” International Journal of Financial Studies, vol. 1, 2013, pp. 62-80. doi: 10.3390/ijfs1030062.
[16]Z. Yang, L. Jin, and M. Wang, “Forecasting baltic panamax index with support vector machine,” Journal of Transportation Systems Engineering and Information Technology, vol. 11, 2011, pp. 50-57. doi:
[17]X. Yuwei, Y. Shaoqiang, F. Yonghui, and Y. Hualong, “Leverage effect analysis of baltic dry index based on EGARCH model,” Journal of Chemical and Pharmaceutical Research, vol. 6, 2014, pp. 289-294.
[18]C.J. Willmott, and K. Matsuura, “Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance,” Clim Res, vol. 30, 2005, pp. 79–82. doi: 10.3354/cr030079.
[19]D.L.J. Alexander, A. Tropsha, and D.A. Winkler, “Beware of R2: simple, unambiguous assessment of the prediction accuracy of QSAR and QSPR models,” J Chem Inf Model, vol. 55, 2015, pp. 1316–1322. doi: 10.1021/acs.jcim.5b00206.