论文部分内容阅读
目的探讨应用ARIMA、季节趋势等时间系列模型预测甲型病毒性肝炎发病率。方法基于2005—2015年宜昌市逐月甲型病毒性肝炎发病率建立两种模型,对2016年甲型病毒性肝炎的发病率进行预测;采用平均绝对百分误差(MAPE)对预测模型进行拟合效果评价。结果两种时间序列与季节趋势模型能较好的模拟甲肝发病,但时间序列模型优于季节性趋势。应用最优模型对2016年甲型病毒性肝炎的发病率进行预测。结论两种时间序列与季节趋势模型能较好的模拟宜昌市甲型病毒性肝炎发病在时间序列的变化趋势,但ARIMA(0,0,1)(0,1,1)12模型能较好的模拟甲型病毒性肝炎发病在时间序列的变化趋势,为制定科学的防控措施和策略提供依据。
Objective To investigate the prediction of the incidence of hepatitis A virus using time series models such as ARIMA and seasonal trend. Methods Based on the monthly incidence of hepatitis A virus in Yichang city from 2005 to 2015, two models were established to predict the incidence of hepatitis A virus in 2016. The prediction model was estimated by using mean absolute percentage error (MAPE) Effect evaluation. Results Two time series and seasonal trend model can simulate the incidence of hepatitis A, but the time series model is better than the seasonal trend. The optimal model was used to predict the incidence of type A hepatitis in 2016. Conclusion Both time series and seasonal trend model can better simulate the trend of time-series changes of hepatitis A virus in Yichang City, but ARIMA (0,0,1) (0,1,1) 12 model is better Of simulated hepatitis A virus in the time series of changes in the trend for the development of scientific prevention and control measures and strategies to provide the basis.