论文部分内容阅读
目的应用季节乘积求和自回归移动平均模型分析南通市水痘每月发病数时间序列,为建立疫情预测模型提供参考依据。方法利用Eviews软件对南通市2008至2014年间水痘病例月报告数进行建模,首先采用自然对数转换、差分法平稳序列,然后估计模型参数,对模型进行检验,筛选最优模型,最后进行预测分析。结果成功建立模型ARIMA(1,0,0)(1,1,0)12,模型表达式为:(1-B12)lnxt(1-0.579 4B)(1+0.312 2B12)=t。模型通过残差白噪声检验,P>0.05;平均绝对百分误差为8.90%,预测精度较高。运用该模型预测2015年1—6月病例数分别为67,36,64,55,72和87例。结论该模型对南通市水痘发病情况拟合和趋势预测结果较好,可根据预测结果开展水痘疫情相关防控准备工作。
Objective To analyze the time series of monthly incidence of chickenpox in Nantong City using the seasonal product autoregression moving average model and provide a reference for establishing the epidemic prediction model. Methods Using Eviews software to model the monthly reports of chickenpox in Nantong City from 2008 to 2014, the natural logarithm transformation and difference method were used to stabilize the sequence. Then the model parameters were estimated, the model was tested, the optimal model was screened, and finally the prediction was made analysis. Results The model ARIMA (1,0,0) (1,1,0) 12 was successfully established. The model expression is: (1-B12) lnxt (1-0.579 4B) (1 + 0.312 2B12) = t. The model passed the residual white noise test, P> 0.05; the average absolute percentage error was 8.90%, and the prediction accuracy was high. The model was used to predict the number of cases from January to June 2015 to be 67, 36, 64, 55, 72 and 87 respectively. Conclusion The model predicts the goodness of fit and trend of the occurrence of chickenpox in Nantong. The prediction and control of varicella-related diseases can be carried out.