三种模型在江西省流感样病例预测中的应用与比较
|
摘要
目的构建江西省流感流行趋势最优预测模型,为流感防控提供科学指导。方法从"中国流感监测信息系统"导出江西省2013-2017年每月流感哨点监测数据,并采用自回归(autoregressive,AR)、指数平滑(exponential smoothing,ES)和自回归积分滑动平均(autoregressive integrated moving average,ARIMA)等不同预测方法建模,并将2017年1～12月的预测值和实际比较。结果三种模型的R2分别为0. 731、0. 751和0. 815;均方根误差(root mean square error,RMSE)分别为0. 253、0. 243和0. 212;平均绝对误差(mean absolute error,MAE)分别为0. 189、0. 178和0. 151;平均绝对百分误差(mean absolute percent error,MAPE)分别为10. 092、9. 523和8. 124;平均相对误差(mean relative error,MRE)分别为11. 45%、10. 92%和8. 96%。结论在进行江西省流感样病例就诊百分比趋势建模中,ARIMA是一个较好预测流感样病例就诊百分比的模型。
Objective To establish the optimal epidemical trend prediction model of influenza in Jiangxi Province and provide scientific guidance for influenza prevention and control. Methods Monthly influenza sentinel surveillance data of Jiangxi Province were derived from the "Influenza Surveillance Information System In China"from 2013 to 2017,and the different forecasting methods were used to build model,such as autoregressive( AR),exponential smoothing( ES) and autoregressive integrated moving average( ARIMA),also compared predictions with actual values in 2017. Results R square of the three models were 0. 731,0. 751 and 0. 815 respectively; the root mean square error( MRSE) were0. 253,0. 243 and 0. 212,respectively; mean absolute error( MAE) were 0. 189,0. 178 and 0. 151,respectively; mean absolute percentage error( MAPE) were 10. 092,9. 523 and 8. 124 respectively; the average relative error( MRE) were 11. 45%,10. 92% and 8. 96%,respectively. Conclusions ARIMA was a good model for predicting the percentage of influenza-like illness in outpatient visits in Jiangxi Province.
Objective To establish the optimal epidemical trend prediction model of influenza in Jiangxi Province and provide scientific guidance for influenza prevention and control. Methods Monthly influenza sentinel surveillance data of Jiangxi Province were derived from the "Influenza Surveillance Information System In China"from 2013 to 2017,and the different forecasting methods were used to build model,such as autoregressive( AR),exponential smoothing( ES) and autoregressive integrated moving average( ARIMA),also compared predictions with actual values in 2017. Results R square of the three models were 0. 731,0. 751 and 0. 815 respectively; the root mean square error( MRSE) were0. 253,0. 243 and 0. 212,respectively; mean absolute error( MAE) were 0. 189,0. 178 and 0. 151,respectively; mean absolute percentage error( MAPE) were 10. 092,9. 523 and 8. 124 respectively; the average relative error( MRE) were 11. 45%,10. 92% and 8. 96%,respectively. Conclusions ARIMA was a good model for predicting the percentage of influenza-like illness in outpatient visits in Jiangxi Province.
引文
[1]郭元吉,唐耀武.加强监测迎接流行性感冒新挑战[J].中华流行病学杂志,2003,24(1):2-3.DOI:10.3760/j.issn:0254-6450.2003.01.002.Guo YJ,Tang YW.Strengthen surveillance to meet the new challenges of influenza[J].Chin J Epidemiol,2003,24(1):2-3.DOI:10.3760/j.issn:0254-6450.2003.01.002.
[2]曾光.流感大流行与应对策略分析[J].中华流行病学杂志,2006,27(2):93-95.DOI:10.3760/j.issn:0254-6450.2006.02.001.Zeng G.Situation and strategies on influenza pandemics[J].Chin J Epidemiol,2006,27(2):93-95.DOI:10.3760/j.issn:0254-6450.2006.02.001.
[3]王永怡,李军,张玲霞.关注新发传染病,攻坚H7N9[J].传染病信息,2013,26(2):68.DOI:10.3969/j.issn.1007-8134.2013.02.002.Wang YY,Li J,Zhang LX.Focus on emerging infectious diseases and conquer influenza A(H7N9)virus[J].Infect Dis Info,2013,26(2):68-68.DOI:10.3969/j.issn.1007-8134.2013.02.002.
[4]曾祥兴,李康生.流感百年:20世纪流感大流行的回顾与启示[J].医学与社会,2010,23(11):4-6.DOI:10.3870/YXYSH.2010.11.002.Zeng XX,Li KS.The pandemic history of influenza:review and revelation of the influenza pandemic in last century[J].Medicine and Society,2010,23(11):4-6.DOI:10.3870/YXYSH.2010.11.002.
[5]陈涛,杨静,汪立杰,等.2016-2017年度中国大陆流行性感冒监测分析[J].中国人兽共患病学报,2018,34(3):193-199.DOI:10.3969/j.issn.1002-2694.2018.00.036.Chen T,Yang J,Wang LJ,et al.Surveillance analysis of influenza in Mainland China during 2016-2017 season[J].Chin J Zoono,2018,34(3):193-199.DOI:10.3969/j.issn.1002-2694.2018.00.036.
[6]Shumway RH,Stoffer DS.Time series regression and exploratory data analysis[M].Springer Texts in Statistics,2011:43-82.
[7]于林凤,吴静,周锁兰,等.ARIMA季节模型在我国丙肝发病预测中的应用[J].郑州大学学报(医学版),2014,49(3):344-348.DOI:10.13705/j.issn.1671-6825.2014.03.014.Yu LF,Wu J,Zhou SL,et al.Application of seasonal ARIMAmodel in forecasting incidence of hepatitis C in China[J].JZhengzhou Univ Med Sci,2014,49(3):366-367.DOI:10.13705/j.issn.1671-6825.2014.03.014.
[8]妥小青,张占林,龚政,等.基于ARIMAX模型的乌鲁木齐市流感样病例预测分析[J].中华疾病控制杂志,2018,22(6):590-593.DOI:10.16462/j.cnki.zhjbkz.2018.06.012.Tuo XQ,Zhang ZL,Gong Z,et al.Forecasting influenza like illness in Urumqi based on ARIMAX model[J].Chin J Dis Control Prev,2018,22(6):590-593.DOI:10.16462/j.cnki.zhjbkz.2018.06.012.
[9]陈涛,杨静,汪立杰,等.2015年中国大陆流行性感冒流行特征分析[J].热带病与寄生虫学,2016,14(1):3-5.DOI:10.3969/j.issn.1672-2302.2016.01.002.Chen T,Yang J,Wang LJ,et al.Analysis on the epidemiological characteristics of influenza in mainland China in 2015[J].J Trop Dis Parasitol,2016,14(1):3-5.DOI:10.3969/j.issn.1672-2302.2016.01.002.
[10]蒋德勇,黄星辉,闫姝丽,等.甲型H1N1流感疫苗接种对其暴发疫情的影响[J].职业与健康,2012,28(12):1404,1537.DOI:10.13329/j.cnki.zyyjk.2012.12.021.Jiang DY,Huang XH,Yan SL,et al.Effect of influenza A(H1N1)vaccine against influenza outbreak[J].Occup and Health,2012,28(12):1404,1537.DOI:10.13329/j.cnki.zyyjk.2012.12.021.
[11]葛爱华,鲍倡俊,祁贤,等.江苏省2009-2012年流感监测结果分析[J].中华疾病控制杂志,2014,18(2):123-126.Ge AH,Bao CJ,Qi X,et al.Analysis on influenza surveillance data in Jiangsu Province,2009-2012[J].Chin J Dis Control Prev,2014,18(2):123-126.
[12]张彦琦,黄彦,田考聪.SPSS在医院统计预测中的应用[J].中国医院统计,2002,9(3):131-134.DOI:10.3969/j.issn.1006-5253.2002.03.001.Zhang YQ,Huang Y,Tian KC.Application of SPSS in hospital statistical prediction[J].Chinese Journal of Hospital Statistics,2002,9(3):131-134.DOI:10.3969/j.issn.1006-5253.2002.03.001.
[13]李琼芬,黄甜,王荣华,等.传染病疫情预测预警模型研究进展[J].中国公共卫生,2013,29(11):1695-1697.DOI:10.11847/zgggws2013-29-11-42.Li XF,Huang T,Wang RH,et al.Progress of the models to prediction and early warning of infectious diseases[J].Chin J Public Health,2013,29(11):1695-1697.DOI:10.11847/zgggws2013-29-11-42.
[14]林玫,李永红,董柏青.传染病预测预警方法在我国的应用现状[J].中国热带医学,2010,10(3):308-309.DOI:10.13604/j.cnki.46-1064/r.2010.03.009.Lin M,Li YH,Dong BQ.Current status of application of predicting and preparedness of communicable diseases in China[J].China Tropical Medicine,2010,10(3):308-309.DOI:10.13604/j.cnki.46-1064/r.2010.03.009.
[15]张蔚,张彦琦,杨旭.时间序列资料ARIMA季节乘积模型及其应用[J].第三军医大学学报,2002,24(8):955-957.DOI:10.3321/j.issn:1000-5404.2002.08.026.Zhang W,Zhang YQ,Yang X.Model of multiple seasonal ARIMAand its application to data in time series[J].J Third Mil Med Univ,2002,24(8):955-957.DOI:10.3321/j.issn:1000-5404.2002.08.026.
[16]吴家兵,叶临湘,尤尔科.时间序列模型在传染病发病率预测中的应用[J].中国卫生统计,2006,23(3):276-276.DOI:10.3969/j.issn.1002-3674.2006.03.037.Wu JB,Ye LX,You EK.Application of time series model in the prediction of incidence of infectious diseases[J].Chin J Health Statis,2006,23(3):276-276.DOI:10.3969/j.issn.1002-3674.2006.03.037.
[17]Levine RS,Foster JE,Fullilove RE,et al.Black-white inequalities in mortality and life expectancy,1933-1999:implications for healthy people 2010[J].Public Health Rep,2001,116(5):474-474.DOI:10.1093/phr/116.5.474.
[18]Tong S,Hu W.Climate variation and incidence of Ross river virus in Cairns,Australia:a time-series analysis[J].Environ Health Perspect,2001,109(12):1271-1273.DOI:10.1289/ehp.011091271.
[19]孟蕾,王新华,白亚娜,等.甘肃省哨点医院流感样病例ARIMA模型预测[J].中国公共卫生,2014,30(2):228-230.DOI:10.11847/zgggws2014-30-02-32.Meng L,Wang XH,Bai YN,et al.ARIMA model in prediction of influenza-like illness of sentinel hospitals in Gansu Province[J].Chin J Public Health,2014,30(2):228-230.DOI:10.11847/zgggws2014-30-02-32.
[20]陈超,田鑫,周剑惠,等.流行性腮腺炎疫情时间序列模型建立与应用[J].中国公共卫生,2012,28(2):252-253.DOI:10.11847/zgggws-2012-28-02-61.Cheng C,Tian X,Zhou JH,et al.Establishment and application of time series model in epidemic mumps[J].Chin J Public Health,2012,28(2):252-253.DOI:10.11847/zgggws-2012-28-02-61.
[21]沈忠周,王雅文,马帅,等.ARIMA模型在预测我国乙肝月报告发病数中的应用[J].中华疾病控制杂志,2018,22(3):298-301,307.DOI:10.16462/j.cnki.zhjbkz.2018.03.019.Shen ZZ,Wang YW,Ma S,et al.Prediction on the number of hepatitis B by ARIMA model in China[J].Chin J Dis Control Prev,2018,22(3):298-301,307.DOI:10.16462/j.cnki.zhjbkz.2018.03.019.
[22]周美兰,周志华,罗美玲,等.湖南省哨点医院流感样病例SA-RIMA模型预测[J].实用预防医学,2018,25(3):370-373.DOI:10.3969/j.issn.1006-3110.2018.03.035.Zhou ML,Zhou ZH,Luo ML,et al.Prediction of influenza-like illness in sentinel hospitals in Hunan Province by SARIMA model[J].Pract Prev Med,2018,25(3):370-373.DOI:10.3969/j.issn.1006-3110.2018.03.035.
[23]漆莉,李革,李勤.ARIMA模型在流行性感冒预测中的应用[J].第三军医大学学报,2007,29(3):267-269.DOI:10.16016/j.1000-5404.2007.03.031.Qi L,Li G,Li Q.Applications of ARIMA model on predictive incidence of influenza[J].J Third Mil Med Univ,2007,29(3):267-269.DOI:10.16016/j.1000-5404.2007.03.031.
[24]桂腾叶,陈硕,隗立志,等.基于ARIMA的传染病流行趋势预测及防治对策[J].电子科技,2015,28(12):48-51.DOI:10.16180/j.cnki.issn1007-7820.2015.12.013.Gui TY,Chen S,Wei LZ,et al.Prediction and countermeasures of infectious diseases based on ARIMA[J].Electronic Sci&Tech,2015,28(12):48-51.DOI:10.16180/j.cnki.issn1007-7820.2015.12.013.
[25]Wallis KF,Whitley JD.Sources of error in forecasts and expectations:U.K.economic models,1984-8[J].J Forecast,1991,10(3):231-253.DOI:10.1002/for.3980100302.
[2]曾光.流感大流行与应对策略分析[J].中华流行病学杂志,2006,27(2):93-95.DOI:10.3760/j.issn:0254-6450.2006.02.001.Zeng G.Situation and strategies on influenza pandemics[J].Chin J Epidemiol,2006,27(2):93-95.DOI:10.3760/j.issn:0254-6450.2006.02.001.
[3]王永怡,李军,张玲霞.关注新发传染病,攻坚H7N9[J].传染病信息,2013,26(2):68.DOI:10.3969/j.issn.1007-8134.2013.02.002.Wang YY,Li J,Zhang LX.Focus on emerging infectious diseases and conquer influenza A(H7N9)virus[J].Infect Dis Info,2013,26(2):68-68.DOI:10.3969/j.issn.1007-8134.2013.02.002.
[4]曾祥兴,李康生.流感百年:20世纪流感大流行的回顾与启示[J].医学与社会,2010,23(11):4-6.DOI:10.3870/YXYSH.2010.11.002.Zeng XX,Li KS.The pandemic history of influenza:review and revelation of the influenza pandemic in last century[J].Medicine and Society,2010,23(11):4-6.DOI:10.3870/YXYSH.2010.11.002.
[5]陈涛,杨静,汪立杰,等.2016-2017年度中国大陆流行性感冒监测分析[J].中国人兽共患病学报,2018,34(3):193-199.DOI:10.3969/j.issn.1002-2694.2018.00.036.Chen T,Yang J,Wang LJ,et al.Surveillance analysis of influenza in Mainland China during 2016-2017 season[J].Chin J Zoono,2018,34(3):193-199.DOI:10.3969/j.issn.1002-2694.2018.00.036.
[6]Shumway RH,Stoffer DS.Time series regression and exploratory data analysis[M].Springer Texts in Statistics,2011:43-82.
[7]于林凤,吴静,周锁兰,等.ARIMA季节模型在我国丙肝发病预测中的应用[J].郑州大学学报(医学版),2014,49(3):344-348.DOI:10.13705/j.issn.1671-6825.2014.03.014.Yu LF,Wu J,Zhou SL,et al.Application of seasonal ARIMAmodel in forecasting incidence of hepatitis C in China[J].JZhengzhou Univ Med Sci,2014,49(3):366-367.DOI:10.13705/j.issn.1671-6825.2014.03.014.
[8]妥小青,张占林,龚政,等.基于ARIMAX模型的乌鲁木齐市流感样病例预测分析[J].中华疾病控制杂志,2018,22(6):590-593.DOI:10.16462/j.cnki.zhjbkz.2018.06.012.Tuo XQ,Zhang ZL,Gong Z,et al.Forecasting influenza like illness in Urumqi based on ARIMAX model[J].Chin J Dis Control Prev,2018,22(6):590-593.DOI:10.16462/j.cnki.zhjbkz.2018.06.012.
[9]陈涛,杨静,汪立杰,等.2015年中国大陆流行性感冒流行特征分析[J].热带病与寄生虫学,2016,14(1):3-5.DOI:10.3969/j.issn.1672-2302.2016.01.002.Chen T,Yang J,Wang LJ,et al.Analysis on the epidemiological characteristics of influenza in mainland China in 2015[J].J Trop Dis Parasitol,2016,14(1):3-5.DOI:10.3969/j.issn.1672-2302.2016.01.002.
[10]蒋德勇,黄星辉,闫姝丽,等.甲型H1N1流感疫苗接种对其暴发疫情的影响[J].职业与健康,2012,28(12):1404,1537.DOI:10.13329/j.cnki.zyyjk.2012.12.021.Jiang DY,Huang XH,Yan SL,et al.Effect of influenza A(H1N1)vaccine against influenza outbreak[J].Occup and Health,2012,28(12):1404,1537.DOI:10.13329/j.cnki.zyyjk.2012.12.021.
[11]葛爱华,鲍倡俊,祁贤,等.江苏省2009-2012年流感监测结果分析[J].中华疾病控制杂志,2014,18(2):123-126.Ge AH,Bao CJ,Qi X,et al.Analysis on influenza surveillance data in Jiangsu Province,2009-2012[J].Chin J Dis Control Prev,2014,18(2):123-126.
[12]张彦琦,黄彦,田考聪.SPSS在医院统计预测中的应用[J].中国医院统计,2002,9(3):131-134.DOI:10.3969/j.issn.1006-5253.2002.03.001.Zhang YQ,Huang Y,Tian KC.Application of SPSS in hospital statistical prediction[J].Chinese Journal of Hospital Statistics,2002,9(3):131-134.DOI:10.3969/j.issn.1006-5253.2002.03.001.
[13]李琼芬,黄甜,王荣华,等.传染病疫情预测预警模型研究进展[J].中国公共卫生,2013,29(11):1695-1697.DOI:10.11847/zgggws2013-29-11-42.Li XF,Huang T,Wang RH,et al.Progress of the models to prediction and early warning of infectious diseases[J].Chin J Public Health,2013,29(11):1695-1697.DOI:10.11847/zgggws2013-29-11-42.
[14]林玫,李永红,董柏青.传染病预测预警方法在我国的应用现状[J].中国热带医学,2010,10(3):308-309.DOI:10.13604/j.cnki.46-1064/r.2010.03.009.Lin M,Li YH,Dong BQ.Current status of application of predicting and preparedness of communicable diseases in China[J].China Tropical Medicine,2010,10(3):308-309.DOI:10.13604/j.cnki.46-1064/r.2010.03.009.
[15]张蔚,张彦琦,杨旭.时间序列资料ARIMA季节乘积模型及其应用[J].第三军医大学学报,2002,24(8):955-957.DOI:10.3321/j.issn:1000-5404.2002.08.026.Zhang W,Zhang YQ,Yang X.Model of multiple seasonal ARIMAand its application to data in time series[J].J Third Mil Med Univ,2002,24(8):955-957.DOI:10.3321/j.issn:1000-5404.2002.08.026.
[16]吴家兵,叶临湘,尤尔科.时间序列模型在传染病发病率预测中的应用[J].中国卫生统计,2006,23(3):276-276.DOI:10.3969/j.issn.1002-3674.2006.03.037.Wu JB,Ye LX,You EK.Application of time series model in the prediction of incidence of infectious diseases[J].Chin J Health Statis,2006,23(3):276-276.DOI:10.3969/j.issn.1002-3674.2006.03.037.
[17]Levine RS,Foster JE,Fullilove RE,et al.Black-white inequalities in mortality and life expectancy,1933-1999:implications for healthy people 2010[J].Public Health Rep,2001,116(5):474-474.DOI:10.1093/phr/116.5.474.
[18]Tong S,Hu W.Climate variation and incidence of Ross river virus in Cairns,Australia:a time-series analysis[J].Environ Health Perspect,2001,109(12):1271-1273.DOI:10.1289/ehp.011091271.
[19]孟蕾,王新华,白亚娜,等.甘肃省哨点医院流感样病例ARIMA模型预测[J].中国公共卫生,2014,30(2):228-230.DOI:10.11847/zgggws2014-30-02-32.Meng L,Wang XH,Bai YN,et al.ARIMA model in prediction of influenza-like illness of sentinel hospitals in Gansu Province[J].Chin J Public Health,2014,30(2):228-230.DOI:10.11847/zgggws2014-30-02-32.
[20]陈超,田鑫,周剑惠,等.流行性腮腺炎疫情时间序列模型建立与应用[J].中国公共卫生,2012,28(2):252-253.DOI:10.11847/zgggws-2012-28-02-61.Cheng C,Tian X,Zhou JH,et al.Establishment and application of time series model in epidemic mumps[J].Chin J Public Health,2012,28(2):252-253.DOI:10.11847/zgggws-2012-28-02-61.
[21]沈忠周,王雅文,马帅,等.ARIMA模型在预测我国乙肝月报告发病数中的应用[J].中华疾病控制杂志,2018,22(3):298-301,307.DOI:10.16462/j.cnki.zhjbkz.2018.03.019.Shen ZZ,Wang YW,Ma S,et al.Prediction on the number of hepatitis B by ARIMA model in China[J].Chin J Dis Control Prev,2018,22(3):298-301,307.DOI:10.16462/j.cnki.zhjbkz.2018.03.019.
[22]周美兰,周志华,罗美玲,等.湖南省哨点医院流感样病例SA-RIMA模型预测[J].实用预防医学,2018,25(3):370-373.DOI:10.3969/j.issn.1006-3110.2018.03.035.Zhou ML,Zhou ZH,Luo ML,et al.Prediction of influenza-like illness in sentinel hospitals in Hunan Province by SARIMA model[J].Pract Prev Med,2018,25(3):370-373.DOI:10.3969/j.issn.1006-3110.2018.03.035.
[23]漆莉,李革,李勤.ARIMA模型在流行性感冒预测中的应用[J].第三军医大学学报,2007,29(3):267-269.DOI:10.16016/j.1000-5404.2007.03.031.Qi L,Li G,Li Q.Applications of ARIMA model on predictive incidence of influenza[J].J Third Mil Med Univ,2007,29(3):267-269.DOI:10.16016/j.1000-5404.2007.03.031.
[24]桂腾叶,陈硕,隗立志,等.基于ARIMA的传染病流行趋势预测及防治对策[J].电子科技,2015,28(12):48-51.DOI:10.16180/j.cnki.issn1007-7820.2015.12.013.Gui TY,Chen S,Wei LZ,et al.Prediction and countermeasures of infectious diseases based on ARIMA[J].Electronic Sci&Tech,2015,28(12):48-51.DOI:10.16180/j.cnki.issn1007-7820.2015.12.013.
[25]Wallis KF,Whitley JD.Sources of error in forecasts and expectations:U.K.economic models,1984-8[J].J Forecast,1991,10(3):231-253.DOI:10.1002/for.3980100302.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |