刘峰,朱妮,邱琳,王敬军,王维华.ARIMA乘积季节模型在陕西省手足口病预测中的应用[J].Chinese journal of Epidemiology,2016,37(8):1117-1120 |
ARIMA乘积季节模型在陕西省手足口病预测中的应用 |
Application of R-based multiple seasonal ARIMA model, in predicting the incidence of hand, foot and mouth disease in Shaanxi province |
Received:March 11, 2016 |
DOI:10.3760/cma.j.issn.0254-6450.2016.08.013 |
KeyWord: 手足口病 自回归求和移动平均乘积季节模型 预测 |
English Key Word: Hand foot and mouth disease Multiple seasonal ARIMA model Prediction |
FundProject: |
Author Name | Affiliation | E-mail | Liu Feng | Shaanxi Provincal Center for Disease Control and Prevention, Xi'an 710054, China | | Zhu Ni | Institute for Health Information of Shaanxi Provincial Center for Disease Control and Prevention, Xi'an 710054, China | | Qiu Lin | Institute for Health Information of Shaanxi Provincial Center for Disease Control and Prevention, Xi'an 710054, China | | Wang Jingjun | Shaanxi Provincal Center for Disease Control and Prevention, Xi'an 710054, China | | Wang Weihua | Institute for Communicable Disease Control and Prevention of Shaanxi Provincial Center for Disease Control and Prevention, Xi'an 710054, China | myjshy@163.com |
|
Hits: 4254 |
Download times: 2749 |
Abstract: |
目的 应用自回归求和移动平均(ARIMA)乘积季节模型预测陕西省手足口病发病人数。方法 利用R软件对2009年1月至2015年6月手足口病数据构建ARIMA乘积季节模型,同时利用2015年7-12月实际发病人数与模型拟合数据比较,评价模型的预测性能,预测2016-2017年的手足口病发病人数。结果 陕西省手足口病发病人数具有明显的季节性,建立了ARIMA(2,1,0)×(1,1,0)12乘积季节模型,预测方程为(1-B)(1-B12)Ln(Xt)=((1-1.000B)/((1-0.532B-0.363B2)(1-0.644B12-0.454B122)))εt,与2015年7-12月实际发病人数比较,绝对误差的平均值为531.535,相对误差的平均值为0.114,预测2016-2017年陕西省手足口病发病人数与2015年基本保持一致。结论 ARIMA(2,1,0)×(1,1,0)12乘积季节模型可用于陕西省手足口病的发病人数预测。 |
English Abstract: |
Objective To apply the 'auto-regressive integrated moving average product seasonal model' in predicting the number of hand, foot and mouth disease in Shaanxi province. Methods In Shaanxi province, the trend of hand, foot and mouth disease was analyzed and tested, under the use of R software, between January 2009 and June 2015. Multiple seasonal ARIMA model was then fitted under time series to predict the number of hand, foot and mouth disease in 2016 and 2017. Results Seasonal effect was seen in hand, foot and mouth disease in Shaanxi province. A multiple seasonal ARIMA (2,1,0)×(1,1,0)12 was established, with the equation as (1-B)(1-B12)Ln(Xt)=((1-1.000B)/((1-0.532B-0.363B2)(1-0.644B12-0.454B122)))εt. The mean of absolute error and the relative error were 531.535 and 0.114, respectively when compared to the simulated number of patients from Jun to Dec in 2015. Results under the prediction of multiple seasonal ARIMA model showed that the numbers of patients in both 2016 and 2017 were similar to that of 2015 in Shaanxi province. Conclusion Multiple seasonal ARIMA (2,1,0)×(1,1,0)12 model could be used to successfully predict the incidence of hand, foot and mouth disease in Shaanxi province. |
View Fulltext
Html FullText
View/Add Comment Download reader |
Close |
|
|
|