基于自回归移动平均模型的浙江省肺结核发病趋势预测

茅蓉 王远航 葛锐

茅蓉, 王远航, 葛锐. 基于自回归移动平均模型的浙江省肺结核发病趋势预测[J]. 疾病监测, 2022, 37(5): 652-656. doi: 10.3784/jbjc.202109240520
引用本文: 茅蓉, 王远航, 葛锐. 基于自回归移动平均模型的浙江省肺结核发病趋势预测[J]. 疾病监测, 2022, 37(5): 652-656. doi: 10.3784/jbjc.202109240520
Mao Rong, Wang Yuanhang, Ge Rui. Prediction of incidence of pulmonary tuberculosis in Zhejiang based on autoregressive integrated moving average model[J]. Disease Surveillance, 2022, 37(5): 652-656. doi: 10.3784/jbjc.202109240520
Citation: Mao Rong, Wang Yuanhang, Ge Rui. Prediction of incidence of pulmonary tuberculosis in Zhejiang based on autoregressive integrated moving average model[J]. Disease Surveillance, 2022, 37(5): 652-656. doi: 10.3784/jbjc.202109240520

基于自回归移动平均模型的浙江省肺结核发病趋势预测

doi: 10.3784/jbjc.202109240520
详细信息
    作者简介:

    茅蓉,女,浙江省嘉兴市人,硕士,主任医师,主要从事结核病防治工作,Email:qiuyin401@163.com

    通讯作者:

    王远航,Tel:0573–83685227,Email:wyhjxcdc@163.com

  • 中图分类号: R211; R521

Prediction of incidence of pulmonary tuberculosis in Zhejiang based on autoregressive integrated moving average model

More Information
  • 摘要:   目的   应用自回归移动平均(ARIMA)模型对浙江省肺结核疫情预测分析,为浙江省肺结核精准化防控工作提供科学依据。   方法   收集2011年1月至2021年8月的浙江省肺结核发病率数据, 基于R软件(4.0.3)利用2011 — 2020年肺结核发病率数据建立ARIMA模型,比较2021年1—8月预测数据和实际数据并选择最优模型。   结果   2011年1月至2020年12月浙江省报告新发肺结核患者总计374 718例,呈逐年下降趋势,每年12月至次年2月发病率较低,3 — 5月相对较高。 确定最优模型为ARIMA(2,1,0)(1,1,2)12,该模型拟合的2021年1 — 8月浙江省肺结核发病率预测值与真实值的平均相对误差为8.87%,赤池信息准则值、贝叶斯信息准则值、均方根误差值和平均绝对百分比误差值分别为95.02、111.05、0.30和4.39。  结论   ARIMA(2,1,0)(1,1,2)12模型能较好地拟合预测浙江省肺结核发病率在时间序列上的变动趋势,但需根据实际情况动态调整,提高预测精度。
  • 图  1  2011-2020年浙江省肺结核发病率时间序列

    Figure  1.  Time series of incidence rate of pulmonary TB in Zhejiang, 2011−2020

    图  2  差分后序列的自相关图和偏相关图

    Figure  2.  Diagrams of ACF and PACF after sequence difference

    图  3  ARIMA(2,1,0)(1,1,2)12模型预测浙江省肺结核发病率拟合结果

    注:阴影部分表示预测值均落在95%置信区间内

    Figure  3.  Fitting graph of ARIMa (2,1,0) (1,1,2) 12 model for predicting incidence of pulmonary TB in Zhejiang

    表  1  备选模型参数估计

    Table  1.   Parameters of alternative models

    参数ARIMA(2,1,0)
    (1,1,2)12
    ARIMA(2,1,0)
    (2,1,1)12
    ARIMA(2,1,0)
    (1,1,1)12
    ARIMA(2,1,0)
    (0,1,2)12
    ARIMA(3,0,0)
    (2,1,1)12
    估计值标准误估计值标准误估计值标准误估计值标准误估计值标准误
    AR10.520.090.510.090.510.090.520.090.460.10
    AR20.270.090.250.100.280.090.260.090.240.10
    AR30.230.10
    SAR10.600.120.210.150.270.150.200.15
    SAR20.170.130.170.13
    SMA10.150.200.810.180.220.250.610.210.760.16
    SMA20.850.190.270.17
    AIC 95.02 100.43 100.07 99.08 100.99
    BIC111.05 116.46 113.43 112.44 116.75
    L−B检验P 0.158 0.198 0.222 0.215 0.268
    下载: 导出CSV

    表  2  备选模型精度评价

    Table  2.   Accuracy evaluation of alternative models

     参数ARIMA
    (2,1,0)
    (1,1,2)12
    ARIMA
    (2,1,0)
    (2,1,1)12
    ARIMA
    (2,1,0)
    (1,1,1)12
    ARIMA
    (2,1,0)
    (0,1,2)12
    ARIMA
    (3,0,0)
    (2,1,1)12
     ME0.000.000.000.000.05
     RMSE0.300.330.330.320.32
     MAE0.220.240.240.240.24
     MPE0.410.440.420.431.26
     MAPE4.394.694.744.674.71
     MASE0.450.490.490.480.48
     ACF10.020.020.250.210.00
    下载: 导出CSV

    表  3  备选模型2021年1-8月浙江省肺结核发病率预测值与实际值比较情况

    Table  3.   Comparison of predicted value by alternative models and actual value of pulmonary TB incidence in Zhejiang, January−August, 2021

    月份实际值(/10万)ARIMA
    (2,1,0)
    (1,1,2)12
    ARIMA
    (2,1,0)
    (2,1,1)12
    ARIMA
    (2,1,0)
    (1,1,1)12
    ARIMA
    (2,1,0)
    (0,1,2)12
    ARIMA
    (3,0,0)
    (2,1,1)12
    预测值(/10万)相对误差(%)预测值(/10万)相对误差(%)预测值(/10万)相对误差(%)预测值(/10万)相对误差(%)预测发值(/10万)相对误差(%)
    13.283.42 4.293.37 2.683.41 4.053.39 3.203.46 5.44
    22.692.74 1.642.78 3.202.81 4.552.79 3.502.83 5.22
    34.214.63 9.984.6410.034.6811.044.6410.224.7312.19
    44.254.9516.514.9817.165.0819.515.0418.545.1120.18
    54.374.9313.005.0515.605.1417.745.1016.755.2219.45
    64.364.748.864.8210.694.8711.804.8611.565.0315.47
    74.424.849.474.9812.545.0514.175.0213.605.2117.88
    84.304.627.234.7410.044.7811.074.7610.534.9915.88
    平均相对误差(%)8.8710.2411.7410.9913.96
    下载: 导出CSV
  • [1] 葛均波, 徐永健, 王辰. 内科学[M]. 9版. 北京: 人民卫生出版社, 2018: 62–74.

    Ge JB, Xu YJ, Wang C. Internal medicine[M]. 9th ed. Beijing: People's Medical Publishing House, 2018: 62–74.
    [2] 中华人民共和国疾病预防控制局. 关于印发遏制结核病行动计划(2019-2022年)的通知[EB/OL]. (2019−06−13)[2021−09−17]. http://www.nhc.gov.cn/jkj/s3589/201906/b30ae2842c5e4c9ea2f9d5557ad4b95f.shtml.

    Bureau of Disease Control and Prevention of the People's Republic of China. Notification on the action plan for the suppression of tuberculosis (2019−2022)[EB/OL]. (2019−06−13)[2021−09−17]. http://www.nhc.gov.cn/jkj/s3589/201906/b30ae2842c5e4c9ea2f9d5557ad4b95f.shtml.
    [3] 王前, 李涛, 杜昕, 等. 2015-2019年全国肺结核报告发病情况分析[J]. 中国防痨杂志,2021,43(2):107–112. DOI:10.3969/j.issn.1000−6621.2021.02.002.

    Wang Q, Li T, Du X, et al. The analysis of national tuberculosis reported incidence and mortality, 2015−2019[J]. Chin J Antituberc, 2021, 43(2): 107–112. DOI: 10.3969/j.issn.1000−6621.2021.02.002.
    [4] 丁哲渊, 吴昊澄, 鲁琴宝, 等. 2020年浙江省法定传染病疫情分析[J]. 预防医学,2021,33(4):325–331. DOI:10.19485/j.cnki.issn2096−5087.2021.04.001.

    Ding ZY, Wu HC, Lu QB, et al. Epidemiological characteristics of the notifiable infectious diseases reported in Zhejiang province, 2020[J]. Prev Med, 2021, 33(4): 325–331. DOI: 10.19485/j.cnki.issn2096−5087.2021.04.001.
    [5] 国务院办公厅. “十三五”全国结核病防治规划[EB/OL]. (2017−02−16)[2021−09−17]. http://www.nhc.gov.cn/bgt/gwywj2/201702/7b2a362da2da4841a362f8eb36575b67.shtml.

    General Office of the State Council. 13th Five-Year national tuberculosis prevention and control plan[EB/OL]. (2017−02−16)[2021−09−17]. http://www.nhc.gov.cn/bgt/gwywj2/201702/7b2a362da2da4841a362f8eb36575b67.shtml.
    [6] 言晨绮, 王瑞白, 刘海灿, 等. ARIMA模型预测2018-2019年我国肺结核发病趋势的应用[J]. 中华流行病学杂志,2019,40(6):633–637. DOI:10.3760/cma.j.issn.0254−6450.2019.06.006.

    Yan CQ, Wang RB, Liu HC, et al. Application of ARIMA model in predicting the incidence of tuberculosis in China from 2018 to 2019[J]. Chin J Epidemiol, 2019, 40(6): 633–637. DOI: 10.3760/cma.j.issn.0254−6450.2019.06.006.
    [7] 王晨, 郭倩, 周罗晶. 基于R语言的ARIMA模型对流感样病例发病趋势的预测[J]. 中华疾病控制杂志,2018,22(9):957–960. DOI: 10.16462/j.cnki.zhjbkz.2018.09020.

    Wang C, Guo Q, Zhou LJ. Forecast of incidence trend of influenza-like illness by the ARIMA model based on R[J]. Chin J Dis Control Prev, 2018, 22(9): 957–960. DOI:  10.16462/j.cnki.zhjbkz.2018.09020.
    [8] 雷宇, 何立乾, 张广川, 等. 自回归移动平均模型的建立及在广州市肺结核发病预测中的应用[J]. 中国防痨杂志,2021,43(6):569–575. DOI:10.3969/j.issn.1000−6621.2021.06.009.

    Lei Y, He LQ, Zhang GC, et al. Establishment of ARIMA model and its application on the prediction of pulmonary tuberculosis incidence in Guangzhou[J]. Chin J Antituberc, 2021, 43(6): 569–575. DOI: 10.3969/j.issn.1000−6621.2021.06.009.
    [9] 杨长庆, 嵇冬静, 李峰, 等. ARIMA模型在盐城市肺结核发病率预测中的应用[J]. 南通大学学报:医学版,2017,37(4):331–334. DOI:10.16424/j.cnki.cn32−1807/r.2017.04.011.

    Yang CQ, Ji DJ, Li F, et al. Application of ARIMA model in predicting incidence of pulmonary tuberculosis incidence in Yancheng city[J]. J Nantong University:Med Sci, 2017, 37(4): 331–334. DOI: 10.16424/j.cnki.cn32−1807/r.2017.04.011.
    [10] 秘玉清, 张继萍, 殷延玲, 等. 基于ARIMA模型的山东省肺结核发病趋势预测[J]. 中国卫生统计,2018,35(6):879–881.

    Mi YQ, Zhang JP, Yin YL, et al. Prediction of pulmonary tuberculosis morbidity trend in Shandong province based on ARIMA model[J]. Chin J Health Stat, 2018, 35(6): 879–881.
    [11] 孙娜, 许小珊, 冯佳宁, 等. ARIMA与GM(1, 1)模型对我国肺结核年发病人数预测情况的比较[J]. 中国卫生统计,2019,36(1):71–74.

    Sun N, Xu XS, Feng JN, et al. Comparison of ARIMA and GM (1, 1) models in predicting the morbidity of pulmonary tuberculosis in China[J]. Chin J Health Stat, 2019, 36(1): 71–74.
    [12] 黄国宝, 黎衍云, 吴菲, 等. ARIMA模型和ARIMA-SVM模型对上海市2型糖尿病患者肺结核发病的预测效果[J]. 复旦学报(医学版),2020,47(6):899–905. DOI:10.3969/j.issn.1672−8467.2020.06.016.

    Huang GB, Li YY, Wu F, at al. Forecasting ability of ARIMA model and ARIMA-SVM model for incidence of pulmonary tuberculosis in patients with type 2 diabetes mellitus in Shanghai[J]. Fudan Univ J Med Sci, 2020, 47(6): 899–905. DOI: 10.3969/j.issn.1672−8467.2020.06.016.
    [13] 张蓓蓓, 彭献镇, 王建明, 等. 中国肺结核发病趋势的ARIMA乘积季节模型构建[J]. 江苏预防医学,2021,32(4):400–402,408. DOI:10.13668/j.issn.1006−9070.2021.04.006.

    Zhang BB, Peng XZ, Wang JM, et al. Establishment of tuberculosis incidence trend model in China by multiple seasonal ARIMA model[J]. Jiangsu J Prev Med, 2021, 32(4): 400–402,408. DOI: 10.13668/j.issn.1006−9070.2021.04.006.
    [14] 张顺先, 邱磊, 张少言, 等. ARIMA模型的建立及对中国肺结核月报告例数的预测效果研究[J]. 中国防痨杂志,2020,42(6):614–620. DOI:10.3969/j.issn.1000−6621.2020.06.014.

    Zhang SX, Qiu L, Zhang SY, et al. A study of prediction effect of autoregressive integrated moving average model on the monthly reported pulmonary tuberculosis cases in China[J]. Chin J Antituberc, 2020, 42(6): 614–620. DOI: 10.3969/j.issn.1000−6621.2020.06.014.
    [15] 杨召, 叶中辉, 尤爱国, 等. 乘积季节ARIMA模型在结核病发病率预测中应用[J]. 中国公共卫生,2013,29(4):469–472. DOI:10.11847/zgggws2013−29−04−01.

    Yang Z, Ye ZH, You AG, et al. Application of multiple seasonal ARIMA model in prediction of tuberculosis incidence[J]. Chin J Public Health, 2013, 29(4): 469–472. DOI: 10.11847/zgggws2013−29−04−01.
    [16] 卞子龙, 卓莹莹, 贺志强, 等. 应用乘积季节模型与指数平滑模型预测上海市肺结核疫情[J]. 南京医科大学学报:自然科学版,2021,41(2):268–273. DOI: 10.7655/NYDXBNS20210223.

    Bian ZL, Zhuo YY, He ZQ, et al. Application of multiple seasonal model and exponential smoothing model in predicitng pulmonary tuberculosis epidemic in Shanghai[J]. J Nanjing Med Univ:Nat Sci, 2021, 41(2): 268–273. DOI:  10.7655/NYDXBNS20210223.
    [17] 游楠楠, 刘巧, 李忠奇, 等. 基于ARIMA模型的江苏省不同地区肺结核发病趋势的预测[J]. 南京医科大学学报:自然科学版,2020,40(6):909–914,916. DOI: 10.7655/NYDXBNS20200626.

    You NN, Liu Q, Li ZQ, et al. Forecast of tuberculosis incidence in different regions of Jiangsu province based on ARIMA model[J]. J Nanjing Med Univ:Nat Sci, 2020, 40(6): 909–914,916. DOI:  10.7655/NYDXBNS20200626.
    [18] 原梅, 张治国, 豆智慧, 等. 北京市昌平区肺结核发病数ARIMA模型预测[J]. 疾病监测,2015,30(12):1045–1049. DOI:10.3784/j.issn.1003−9961.2015.12.014.

    Yuan M, Zhang ZG, Dou ZH, et al. Application of ARIMA model in predicting incidence of pulmonary tuberculosis in Changping district, Beijing[J]. Dis Surveill, 2015, 30(12): 1045–1049. DOI: 10.3784/j.issn.1003−9961.2015.12.014.
    [19] 李家琦, 王雷, 宋媛媛, 等. ARIMA模型在湖北省肺结核发病数预测中的应用[J]. 公共卫生与预防医学,2018,29(5):37–40. DOI:10.3969/j.issn.1006−2483.2018.05.010.

    Li JQ, Wang L, Song YY, et al. Application of ARIMA model in prediction of incidence of tuberculosis in Hubei[J]. J Public Health Prev Med, 2018, 29(5): 37–40. DOI: 10.3969/j.issn.1006−2483.2018.05.010.
  • 加载中
图(3) / 表(3)
计量
  • 文章访问数:  199
  • HTML全文浏览量:  98
  • PDF下载量:  35
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-09-24
  • 网络出版日期:  2022-01-18
  • 刊出日期:  2022-05-31

目录

    /

    返回文章
    返回

    在线交流

    防诈骗公告

    近期有不法分子以本刊编辑身份添加作者微信,请务必提高警惕!本刊关于稿件的一切事项通知均采用编辑部唯一邮箱(jbjc@icdc.cn)和座机(010-58900732)联系作者,且在录用稿件后仅收取版面费,无其他任何名目费用(如审稿费和加急费等),非编辑部邮箱发送的本刊收费用通知等均为诈骗,不要随意汇入款项!如有可疑及时致电编辑部核实确认!