Analysis of Network Characteristics of BDI under Dynamic Complex Network

Gailan Liang

《International Core Journal of Engineering》 7卷4期 (2021/04) Pp. 153-163

https://doi.org/10.6919/ICJE.202104_7(4).0022

摘要

To reflect the long-period BDI index time series changes on the dynamic complex network, the article symbolizes the BDI time series data that changes continuously in time into a modal form, obtains the BDI index time fluctuation sequence, and then performs the time scale Coarse-grained processing, construction of time sliding data window to get the BDI index dynamic complex network. By analyzing the dynamic characteristics of the dynamic complex network such as the modal strength and distribution, network structure entropy, weighted clustering coefficient, average path length, betweenness, the general laws and characteristics of the Baltic dry bulk index network are studied .Studies have shown that: the Baltic Dry Index shows periodicity, power law, small-world characteristics, scale-free, and homogeneity. There are some small groups in the modal conversion cycle. The average period of modal conversion is 3~5 months, and there are more cases of slight fluctuations. The research on the characteristics of the Baltic Dry Bulk Index network can provide forecasts and warnings for the shipping market and shipping port and shipping companies.

關鍵字

Waterway Transportation ； Shipping Market ； Dynamic Complex Network ； Dry Bulk Index ； Fluctuation Characteristics

參考文獻

Chien-Chang Chou, Keng-Shou Lin. A fuzzy neural network combined with technical indicators and its application to Baltic Dry Index forecasting. 2019, 18(2):82-91.

Qingcheng Zeng, Chenrui Qu, Adolf K.y. Ng, et al. A new approach for Baltic Dry Index forecasting based on empirical mode decomposition and neural networks. 2016, 18(2):192-210.

Imam Mustafa Kamal, Hyerim Bae, Sim Sunghyun, et al. DERN: Deep Ensemble Learning Model for Short- and Long-Term Prediction of Baltic Dry Index. 2020, 10(4).

X. Zhang, M.Y. Chen, M.G. Wang, Y.E. Ge, H.E. Stanley. A novel hybrid approach to Baltic Dry Index forecasting based on a combined dynamic fluctuation network and artificial intelligence method[J]. Applied Mathematics and Computation,2019,361.

Ding Liu, Cunbin Li, Defu Zhao, et al. Research on the joint fluctuation laws between locational marginal price and renewables based on complex networks: A case study in Independent System Operator New England. 2019, 7(6):2866-2883.