论文部分内容阅读
[目的]探讨应用广义相加模型进行中国2010年上海世博会园区医疗站就诊人数预测的可行性。[方法]采用时间序列的广义相加模型,在控制星期效应的基础上,对2010年5月1日至8月8日开园100d间的每日世博园区就诊人数、入园人数和气象因素资料进行模型拟合。[结果]园区就诊人数存在周末效应,就诊人数随入园总人数的增加而增加;风速对就诊人数的影响趋势是随风速的增加,就诊人数先增加后减少;随着气温上升和日温差增加,就诊人数呈上升趋势(P<0.01)。利用模型对每日就诊人数进行预测,预测的平均相对误差为10.44%。[结论]广义相加模型能较好拟合世博园区医疗站就诊人数的趋势,可用于大型活动中的预测研究。
[Objective] The research aimed to discuss the feasibility of using the generalized additive model to predict the number of medical staffs attending the 2010 Shanghai World Expo site in China. [Method] Based on the generalized additive time series model, on the basis of controlling the weekly effect, the paper analyzed the attendance, attendance and meteorological factors of the daily EXPO Park during the 100 days from May 1, 2010 to August 8, 2010 Model fit. [Results] There was a weekend effect on the number of attendance in the park. The number of attendances increased with the increase of the total number of attendants in the park. The trend of the number of attendants affected by wind speed tended to increase with the increase of wind speed. The number of attendants first increased and then decreased. With the increase of temperature and the difference of daily temperature Increase, the number of visits showed an upward trend (P <0.01). The model was used to predict the number of daily visits and the average relative error was 10.44%. [Conclusion] The generalized additive model can better fit the trend of attendance at the medical station in the Expo site and can be used in the prediction of large-scale activities.