论文部分内容阅读
目的 探讨乘积季节自回归求和滑动平均模型 (integrated autore-gressive moving average model, ARIMA) 在荆州市乙肝发病预测中的应用, 为乙肝预防控制提供参考.方法 利用2004-2015年乙肝网络监测数据对荆州市乙肝发病率数据构建乘积季节ARIMA模型, 同时利用2016年实际发病率与模型拟合数据进行比较, 评价模型的预测性能, 并预测荆州市2017年的乙肝发病率.结果 荆州市乙肝发病率预测最优模型为ARIMA (0, 1, 1) (0, 1, 1) 12模型, 利用2016年拟合值与实际乙肝发病率比较, 相对误差介于1.33％～27.80％之间, 平均相对误差10.23％, 提示ARIMA (0, 1, 1) (0, 1, 1) 12模型具有较佳的预测能力.预测2017年荆州市乙肝疫情与2016年基本一致, 发病整体平稳.结论 ARIMA (0, 1, 1) (0, 1, 1) 12模型可用于荆州市乙肝发病率的预测, 对乙肝预防控制产生积极的指导作用.“,”Objective To explore the application of multiple seasonal autoregressive integrated moving average model (ARIMA) in predicting hepatitis B incidence in Jingzhou and to provide reference for the prevention and control of hepatitis B. Methods Based on the network monitoring data of hepatitis B from 2004 to 2015, an ARIMA model of incidence rate of hepatitis B in Jingzhou City was established. Then the actual incidence rate in 2016 was compared with the fitting data of the model to evaluate its predictive performance and predict the incidence rate of 2017. Results The comparison between the fitting data of 2016 and the actual incidence rate revealed that ARIMA (0, 1, 1) (0, 1, 1) 12 model was the optimal prediction model, with the relative error lying between 1. 33％ and 27. 80％ and the average relative error being 10. 23％. It suggested that the overall incidence was going to be stable and the incidence of hepatitis B in 2017 would be in line with that of 2016. Conclusion ARIMA (0, 1, 1) (0, 1, 1) 12 model is applicable to the prediction of the incidence of hepatitis B in Jingzhou, and it can provide scientific reference for the prevention and control of hepatitis B.