Application of multiple seasonal ARIMA model in forecasting the incidence of brucellosis in Qingzhou
Liu Jie1, Wu Qinfa2, Li Weiguo1, Xiao Yufei3, Mao Qian3 Shi Fuyan3, Wang Suzhen3.
1.Qingzhou Institute of Endemic Disease Prevention and Control, Qingzhou 262500, China;
2.Qingzhou Disease Prevention and Control Center, Qingzhou 262500,China;
3.Department of Health Statistics, School of Public Health and Management, Weifang Medical University, Weifang 261053, China
Abstract:Objective To predict the monthly incidence of brucellosis in Qingzhou by multiple seasonal ARIMA model of autoregressive integrated moving average (ARIMA), and to construct a prediction model for the monthly incidence of brucellosis in Qingzhou so as to provide scientific basis for prevention and control measures of brucellosis in Qingzhou. Methods The monitoring data of brucellosis in Qingzhou from 2011 to 2017 were collected, ARIMA time series model was established with SPSS 25.0 statistical software, the model was tested to predict the monthly incidence of brucellosis in 2018, and the prediction effect of the model was evaluated by the actual monthly incidence of brucellosis in 2018. Results The normalized BIC value of ARIMA(0,1,1)(1,1,0)12model is the smallest, the Box-Ljung test statistic is as follows: Q=23.746, P>0.05, the residual sequence is white noise, thus it is determined to be the optimal model. Conclusion Multiple seasonal ARIMA model is effective in short-term prediction of monthly incidence of brucellosis in Qingzhou.
刘杰,武钦发,李伟国,肖宇飞,毛倩,石福艳,王素珍. ARIMA乘积季节模型在青州市布鲁氏菌病发病预测中的应用[J]. 中国医院统计, 2020, 27(2): 97-100.
Liu Jie1, Wu Qinfa, Li Weiguo, Xiao Yufei, Mao Qian, Shi Fuyan, Wang Suzhen.. Application of multiple seasonal ARIMA model in forecasting the incidence of brucellosis in Qingzhou. journal1, 2020, 27(2): 97-100.