Research on the impact of the epidemic on express delivery volume in Guangdong Province based on a mixed model

This paper aims to explore the impact of the epidemic on the express delivery volume in Guangdong Province. A hybrid model method is used to model and predict the express delivery volume using two algorithms, SARIMA (Seasonal Autoregressive Integrated Moving Average) and SARIMA-GARCH (Generalized Autoregressive Conditional Heteroskedasticity). First, by preprocessing the time series data of the express delivery volume, the SARIMA model is applied to capture the seasonal and trend characteristics of the express delivery volume and identify the non-stationarity of the data. Subsequently, the GARCH model is introduced to solve the heteroskedasticity problem of the residuals, thereby improving the prediction accuracy. By comparing the difference between the predicted value and the actual value, the study found that the demand for express delivery decreased significantly due to the blockade measures and the interruption of the logistics chain in the early stage of the epidemic, but with the adaptation of the market environment and the flexible adjustment of enterprises, the express delivery volume gradually recovered and exceeded the original forecast level.