Research On Compilation And Application Of Shipping Index Based On Complex Network Theory

As a "barometer" of the shipping market,the shipping index is an effective tool to reflect the volatility of the shipping market.It mainly has three economic functions that characterize the shipping freight market,shipping derivatives trading target,and trading scale,and it's an important tool for evading business risks.Building an accurate and effective shipping index plays an important role in water transportation and even the entire economic development.At present,most shipping indexes use the Laspeyres index method.The fixed routes and weights make it difficult for the shipping index to adapt to the rapidly developing shipping market.In order to solve the above problems,this paper proposes a method for building a shipping index based on complex networks.Firstly,a model based on univariate shipping data and multivariable shipping data,with the routes as nodes and the correlation between routes as edges,is used to build a complex network of the shipping market.In order to construct a reasonable network,the Pearson correlation coefficient is used as a measure of the correlation of price fluctuations between routes in different periods,and the distribution of the number of nodes in the maximum connected subgraph under different thresholds is used to determine the optimal threshold,so as to obtain the shipping market networks of different periods.In the research of network structure of the shipping markets,the topological characteristics such as node degree,average agglomeration coefficient and path length are analyzed,and the community filtering structure of the shipping network is analyzed using a clique filtering algorithm.Finally,based on the network characteristics,a new weight distribution and shipping index calculation method is proposed to achieve dynamic weights and indexes.In order to verify the effectiveness of the construction of the shipping market network and the calculation model of the shipping index,empirical analysis is conducted using real data from the Pearl River Shipping Bulk Market.When constructing the network,it was found that only a small part of the relationships between the routes were strong.The number of nodes of the largest connected subgraph is used to determine the correlation coefficient thresholds of different stages and build a multi-stage shipping network.In addition,if the actual route increases or decreases,it only need to adjust the corresponding nodes to construct a new network.In the structural analysis,during the period of market volatility,the Pearl River Shipping Network has higher agglomeration,more complicated network structure,and more obvious characteristics of the small world.According to the analysis results of the community structure,some closely related route groups are found,which provides new ideas for the actual analysis and forecast of changes of routes.By analyzing the new Pearl River Index based on complex networks,it is concluded that the new Pearl River Index can more effectively reflect changes in the shipping market.The new Pearl River Index has a consistent trend with the existing index.The weight distribution based on network characteristics makes the new Pearl River Index more timely and accurate in reflecting market fluctuations,and effectively reflecting the impact of changes the steel,sand,coal,grain,cement and related macro market on the freight rate of the Pearl River system.In summary,the shipping network and shipping index construction method based on the complex network theory can effectively solve the problems of fixed weights and routes in the existing shipping index construction.