Hui-Yu H, Hua-Qin S, Shun-Xian Z, Lin A I, Yan L U, Yu-Chun C, Shi-Zhu L I, Xue-Jiao T, Chun-Li Y, Wei H U, Jia-Xu C
Hebei General Hospital, Hebei Province, Shijiazhuang 050051, China.
National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention; Key Laboratory for Parasitology and Vector Biology, MOH of China, WHO Collaborating Center for Tropical Diseases, National Center for International Research on Tropical Diseases, Shanghai, China.
Zhongguo Xue Xi Chong Bing Fang Zhi Za Zhi. 2017 Aug 15;29(4):436-440. doi: 10.16250/j.32.1374.2017088.
To study the application of autoregressive integrated moving average (ARIMA) model to predict the monthly reported malaria cases in China, so as to provide a reference for prevention and control of malaria. SPSS 24.0 software was used to construct the ARIMA models based on the monthly reported malaria cases of the time series of 20062015 and 2011-2015, respectively. The data of malaria cases from January to December, 2016 were used as validation data to compare the accuracy of the two ARIMA models. The models of the monthly reported cases of malaria in China were ARIMA (2, 1, 1) (1, 1, 0) and ARIMA (1, 0, 0) (1, 1, 0) respectively. The comparison between the predictions of the two models and actual situation of malaria cases showed that the ARIMA model based on the data of 2011-2015 had a higher accuracy of forecasting than the model based on the data of 2006-2015 had. The establishment and prediction of ARIMA model is a dynamic process, which needs to be adjusted unceasingly according to the accumulated data, and in addition, the major changes of epidemic characteristics of infectious diseases must be considered.
研究自回归积分滑动平均（ARIMA）模型在中国疟疾月报告病例预测中的应用，为疟疾防控提供参考。分别使用SPSS 24.0软件基于2006 - 2015年和2011 - 2015年时间序列的疟疾月报告病例构建ARIMA模型。将2016年1月至12月的疟疾病例数据作为验证数据，比较两种ARIMA模型的准确性。中国疟疾月报告病例模型分别为ARIMA(2, 1, 1)(1, 1, 0)和ARIMA(1, 0, 0)(1, 1, 0)。两种模型预测结果与疟疾病例实际情况的比较表明，基于2011 - 2015年数据的ARIMA模型比基于2006 - 2015年数据的模型具有更高的预测准确性。ARIMA模型的建立与预测是一个动态过程，需要根据积累的数据不断调整，此外，还必须考虑传染病流行特征的重大变化。
