| 基于奇异谱分析的ARIMA模型在山西省流感预测中的应用 |
| |
| 引用本文： | 翟梦梦,李国华,高雪芬,王旭春,任浩,李美晨,全帝臣,罗天娥,赵晋芳,陈利民,仇丽霞.基于奇异谱分析的ARIMA模型在山西省流感预测中的应用[J].现代预防医学,2021,0(9):1550-1555. |
| |
| 作者姓名： | 翟梦梦 李国华 高雪芬 王旭春 任浩 李美晨 全帝臣 罗天娥 赵晋芳 陈利民 仇丽霞 |
| |
| 作者单位： | 1.山西医科大学公共卫生学院， 太原 030001；2.山西省疾病预防控制中心；3.山西省人民医院 |
| |
| 摘 要： | 目的 分析基于奇异谱分析（singular spectrum analysis, SSA）的自回归移动平均模型（Autoregressive integrated moving average, ARIMA）模型预测流感样病例 (influenza like illness, ILI) 发病趋势的可行性，为流感防控工作提供合理的预测方法。 方法 利用山西省2010年第14周-2017年第13周的流感监测资料以不同长度配比的训练集、测试集构建SSA-ARIMA模型，并与ARIMA、BP神经网络（Back propagation neural network, BPNN）、广义回归神经网络（General Regression Neural Network, GRNN）模型进行比较。采用平均绝对误差（Mean Absolute Error，MAE）、均方误差（Mean Squared Error，MSE）、均方根误差（Root Mean Squared Error，RMSE）比较各模型预测效果。 结果 模型拟合方面，SSA-ARIMA模型在预测未来一个月发病趋势时的MAE、MSE、RMSE分别为0.163、0.061、0.248；预测六个月时分别为0.161、0.061、0.248；预测一年时分别为0.168、0.066、0.256；均低于ARIMA、BPNN、GRNN。模型预测方面，在预测未来一个月发病趋势时的MAE、MSE、RMSE分别为0.056、0.005、0.068；预测六个月时分别为0.189、0.081、0.285；预测一年时分别为0.210、0.075、0.273；也均低于ARIMA、BPNN、GRNN。 结论 SSA-ARIMA模型对山西省ILI的预测效果优于ARIMA、BPNN、GRNN，可为流感预测提供科学依据。
|
| 关 键 词： | 流感 奇异谱分析 ARIMA BPNN GRNN 预测 |
Application of ARIMA model based on singular spectrum analysis in prediction of influenza in Shanxi Province |
| |
| ZHAI Meng-meng,LI Guo-hua,GAO Xue-fen,WANG Xu-chun,REN Hao,LI Mei-chen,QUAN Di-chen,LUO Tian-e,ZHAO Jin-fang,CHEN Li-min,QIU Li-xia.Application of ARIMA model based on singular spectrum analysis in prediction of influenza in Shanxi Province[J].Modern Preventive Medicine,2021,0(9):1550-1555. |
| |
| Authors: | ZHAI Meng-meng LI Guo-hua GAO Xue-fen WANG Xu-chun REN Hao LI Mei-chen QUAN Di-chen LUO Tian-e ZHAO Jin-fang CHEN Li-min QIU Li-xia |
| |
| Affiliation: | *Shanxi Medical University, Taiyuan, Shanxi 030001, China |
| |
| Abstract: | Abstract: Objective To evaluate the effect of autoregressive moving average (ARIMA) model based on singular spectrum analysis (SSA) in prediction of influenza in Shanxi Province and to provide a reasonable prediction method for influenza prevention and control. Methods The ILI monitoring data from the 14th week of 2010 to the 13th week of 2017 in Shanxi Province were used to establish SSA － ARIMA model with different length matching training sets and test sets，and compared with ARIMA，BP neural network (Back propagation neural network， BPNN) ，and General Regression Neural Network(GRNN) models. Mean Absolute Error (MAE) ，Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) were used to evaluate the prediction effect of the two models. Results In terms of model fitting，SSA － ARIMA model when predicting the incidence trend in the next month were respectively 0. 163，0. 061 and 0. 248; when the prediction was six months， they were respectively 0. 161，0. 061，and 0. 248; when the prediction was one year，they were respectively 0. 168，0. 066，0. 256; the errors were all lower than ARIMA，BPNN，GRNN. In terms of model prediction，the MAE，MSE and RMSE of SSA － ARIMA model when predicting the incidence trend in the next month were respectively 0. 056，0. 005 and0. 068; when the prediction is six months， they were respectively 0. 189，0. 081，and 0. 285; when the prediction is one year，they are respectively 0. 210，0. 075，and 0. 273; the errors also were all lower than ARIMA，BPNN，GRNN. Conclusion The SSA － ARIMA model is better than the ARIMA，BPNN，and GRNN models in predicting ILI in Shanxi Province，and can provide scientific basis for influenza prediction. |
| |
| Keywords: | Influenza SSA ARIMA BPNN GRNN Forecasting |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《现代预防医学》浏览原始摘要信息 |
|
点击此处可从《现代预防医学》下载全文 |
|
