摘要：Based on the time series of the number of ships calling at the container port, this paper takes Shanghai Port and Singapore Port as examples to analyze the statistical characteristics of ship flow time series in container ports. Then, the rescaled range analysis (R/S) method is used to study the long-range dependence (LRD) of ship flow sequence structure in maritime container ports and further analyze the influence of trend component and periodic component on Hurst exponent. The results indicate that both the ship flow time series of Shanghai Port and Singapore Port show the feature of nonlinear and nonstationary from 2013 to 2017, and the data of the two ports have sudden changes in August 2016. They all show specific long-range dependence behavior, but the long-range dependence of ship flow series in Singapore Port is relatively weak. Besides, the trend component and periodic component have significant impact on the Hurst exponent of the time series. Hence the trend component and periodic component should be removed in advance for long-range dependence analysis.
关键词：International Logistics;Container Ports;Time Series Analysis;Rescaled Range Analysis;Long-Range Dependence
