基于ARIMA模型的上海港货物吞吐量和进出口总额预测分析

Forecast and Analysis of Cargo Throughput and Total Import and Export Volume in Shanghai Port Based on ARIMA Model

摘要: 选取上海市1978~2019年外贸进出口总额数据和港口货物吞吐量数据进行分析，本文首先通过构造一元时间序列模型ARIMA(3,2,0)对上海市未来5年外贸进出口总额的变化趋势进行预测；其次，为分析上海市外贸进出口发展对港口货物吞吐量的影响，本文先对二者进行协整检验，并拟合协整动态回归模型和外贸进出口总额对数模型，再根据进出口总额预测值去预测上海市未来5 年港口货物吞吐量的趋势变化。预测结果显示，未来5年的上海市进出口总额和港口货物吞吐量总体上都将呈上升趋势，且通过对比真实值与预测值之间的偏差率发现模型预测的误差率较低，这表明所拟合模型精度较高。

Abstract: This paper analyzes the total import and export volume data and port cargo throughput data of Shanghai from 1978 to 2019. Firstly, it forecasts the change trend of total import and export volume of Shanghai in the next five years by constructing unitary time series model ARIMA (3,2,0). Second-ly, in order to analyze the impact of the development of Shanghai’s foreign trade import and export on the port cargo throughput, this paper first carries out the co-integration test of the two, and fits the co-integration dynamic regression model and the logarithmic model of total foreign trade im-port and export, and then predicts the trend change of Shanghai’s port cargo throughput in the next five years according to the predicted value of total import and export. The forecast results show that in the next five years, the total import and export volume and port cargo throughput of Shanghai will be on the rise, and by comparing the deviation rate between the real value and the predicted value, it is found that the error rate of the model prediction is low, which indicates that the preci-sion of the fitted model is high.