摘要
目的利用SARIMA模型对宁波市手足口病发病情况进行预测,为手足口病防控工作提供依据。方法利用R3.3.2软件对宁波市2012-2015年手足口病的发病数据建立SARIMA模型,利用auto.arima()代码结合模型参数估计和残差诊断选择最优模型,并利用构建的模型对宁波市2016年手足口病发病情况进行预测分析。结果最优模型为SARIMA(1,0,0,)(1,1,0)12,模型的参数ar1和sar1经检验均有统计学意义(P<0.05),残差诊断图显示模型残差为白噪声序列。2016年每月实际发病数均在预测值的95%可信区间内,预测值的变化趋势与实际值的变化趋势基本一致。实际值与预测值误差率波动范围为4.33%~160.36%,其中有4个月份误差率大于100%,5个月份误差率小于20%。结论 SARIMA(1,0,0,)(1,1,0)12模型预测手足口病发病情况有一定的准确性,但需要继续更新相关监测数据并重新构建模型来进一步提高模型预测的准确性。
Objective To predict the incidence of hand-foot-mouth disease by using SARIMA model and provide the evidence for prevention and control. Methods The SARIMA model was established based on the incidence of hand-footmouth disease in Ningbo from 2012 to 2015 using R3.3.2. The best model was selected according to using auto. afima（ ） combined with model parameter estimation and residual diagnosis, which was used to predict the incidence of hand-foot- mouth disease in Ningbo in 2016. Results SARIMA（ 1,0,0, ） （ 1,1,0） 12was selected as the best model ,and its two parameters arl and sarl had statistically significant（ P 〈 0.05 ）. The residual diagnostic graph showed that the model residuals were white noise sequences. The actual value of each month in 2016 were contained in the 95% confidence interval of the predicted value, and the forecast trend and the actual trend were basically same. The deviations of actual values and predict values were between 4.33% - 160.36% ,which four months＇ deviations were larger than 100% and five months＇ deviations were less than 20%. Conclusion The SARIMA（ 1,0,0, ） （ 1,1,0） 12 model has a certain value in prediction of hand-foot-mouth disease incidence. In order to improve the accuracy of the model, monitoring data should be updated continually and the model should be rebuild with new data.
作者
张良
冯伟
李宁
纪威
俞延峰
许国章
ZHANG Liang;FENG Wei;LI Ning;JI Wei;YU Yan-feng;XU Guo-zhang(Center for Disease Control and Prevention of Ningbo,Ningbo ,Zhejiang315010, China;Center for Disease Control and Prevention of Fenghua district in Ningbo)
出处
《中国公共卫生管理》
2018年第2期226-229,共4页
Chinese Journal of Public Health Management
基金
浙江省医学重点学科"现场流行病学"(07-013)
宁波市科学技术局科技创新团队项目(2012B82018)
