基于SARIMA模型的短期机票价格预测

Short-term Airfare Prediction Based on SARIMA Model

为研究不同城市机场间短期机票价格预测的问题,文章将机票数据按照OD、座舱类型、购票时间进行分类,并分场景构建SARIMA模型。采用当前时序的数据标定模型的参数,预测下一时序的机票价格。以2023年7月北京—上海的机票价格数据为例,通过4组实验场景和改进的北京首都—上海虹桥实验场景发现,分场景的SARIMA模型可以较为准确地预测短期机票价格。

In order to study the problem of short-term airfare prediction among airports in different cities,this paper classifies ticket data by OD,cabin type,and ticket purchase time,and constructs a SARIMA model based on different scenarios.It uses the data from the current time series to calibrate the parameters of the model,predict the ticket price for the next time series.Taking the airfare data from Beijing to Shanghai in July 2023 as an example,through four experimental scenarios and an improved Beijing Capital—Shanghai Hongqiao experimental scenario,it is found that the SARIMA model with different scenarios can accurately predict short-term airfare.