论文部分内容阅读
探讨应用时间序列模型对舟山市流感样发病趋势进行预测,为控制流感的流行提供科学依据。对舟山市2009年6月至2015年6月年流感监测哨点医院监测资料分析,应用时间序列拟合模型,并对2015年7~11月进行预测,将预测值与实际值比较。结果显示,确定模型为AR(3),xt=0.619×x(t-1)+0.322×x(t-3)。参数估计结果差异有统计学意义(P<0.05)。预测值的动态趋势与实际趋势较吻合。AR(3)能较好地拟合预测舟山市短期流感样病例发病趋势,可为流感防控提供科学依据。
To explore the application of time series model to predict the trend of influenza-like illness in Zhoushan and provide a scientific basis for controlling the epidemic of influenza. From June 2009 to June 2015, the monitoring data of influenza surveillance sentinel hospitals in Zhoushan City were analyzed, the time series fitting model was applied, and the forecast from July to November in 2015 was made. The predicted value was compared with the actual value. The results show that the model is AR (3), xt = 0.619 × x (t-1) + 0.322 × x (t-3). The difference of parameter estimation results was statistically significant (P <0.05). The dynamic trend of predicted value is in good agreement with the actual trend. AR (3) can well fit and predict the incidence of short-term flu-like illness in Zhoushan City, which can provide a scientific basis for prevention and control of influenza.