基于ARIMA模型的洛阳市手足口病发病率预测
|
摘要
目的建立洛阳市手足口病发病率的自回归积分移动平均(ARIMA)模型,并对洛阳市手足口病的发病率进行预测。方法以2009-2017年洛阳市手足口病发病率数据为基础建立ARIMA模型,并用2018年1月至5月的实际发病率进行验证,评价模型的拟合效果,利用最优模型预测2018年6月至12月洛阳市手足口病发病率。结果在本次研究中建立的ARIMA最优模型为ARIMA(2,0,0)(0,1,1)12,参数均有统计学意义(P <0. 05),拟合优度检验BIC最小为3. 563,残差序列为白噪声(Ljung-Box Q=13. 962,P=0. 528),拟合效果较好。预测出洛阳市2018年6-12月手足口病平均月发病率为13. 16/10万,与2017年同期相比略高。结论 ARIMA(2,0,0)(0,1,1)12模型拟合洛阳市手足口病发病率序列效果较好,可用于在短期上对洛阳市手足口病发病趋势进行预测。
Objective To build an ARIMA model of hand-foot-mouth disease in Luoyang,and to predict the incidence of hand-foot-mouth disease. Methods The ARIMA model was built based on the monthly incidence of hand-foot-mouth disease from 2009 to 2017 in Luoyang. The incidence from January to May in 2018 was used to test the prediction results of the model. Then the optimal model was used to predict the incidence from June to December of 2018. Results The prediction model of hand-foot-mouth disease was ARIMA( 2,0,0)( 0,1,1) 12,and the parameters were all statistically significant( P <0. 05). The minimum goodness of fit value of BIC was 3. 563 and the residual sequence was white noise( Ljung-Box Q =13. 962,P = 0. 528). The goodness of fit was high. The mean monthly prediction results of incidence from June to December in2018 was 13. 16/100000,obviously higher than that in 2017. Conclusion This model can simulate the trend of hand-foot-mouth disease in Luoyang well,and it can be used for the short-term prediction of hand-foot-mouth disease in Luoyang.
Objective To build an ARIMA model of hand-foot-mouth disease in Luoyang,and to predict the incidence of hand-foot-mouth disease. Methods The ARIMA model was built based on the monthly incidence of hand-foot-mouth disease from 2009 to 2017 in Luoyang. The incidence from January to May in 2018 was used to test the prediction results of the model. Then the optimal model was used to predict the incidence from June to December of 2018. Results The prediction model of hand-foot-mouth disease was ARIMA( 2,0,0)( 0,1,1) 12,and the parameters were all statistically significant( P <0. 05). The minimum goodness of fit value of BIC was 3. 563 and the residual sequence was white noise( Ljung-Box Q =13. 962,P = 0. 528). The goodness of fit was high. The mean monthly prediction results of incidence from June to December in2018 was 13. 16/100000,obviously higher than that in 2017. Conclusion This model can simulate the trend of hand-foot-mouth disease in Luoyang well,and it can be used for the short-term prediction of hand-foot-mouth disease in Luoyang.
引文
[1]中华人民共和国中央人民政府.卫生部印发《手足口病预防控制指南(2009版)》[EB/OL].[2018-12-18]. http://www. gov. cn/gzdt/2009-06/04/content_1332078. htm.
[2] Xu W,Zhang MX,Qin EQ,et al. Molecular characterization of wild type measles virus from adult patients in northern China,2014[J].International Journal of Infectious Diseases,2016,45(6):36-42.
[3]陈伟,赵晓静,张杰,等.河南省手足口病的发病时间特征分析[J].中国卫生统计,2016,33(2):212-214.
[4]张倩,陈超.改进的GM(1,1)模型在衡水市乙肝发病率预测中的应用[J].现代预防医学,2017,44(11):1925-1928,1937.
[5]杨仁东,胡世雄,邓志红,等.湖南省手足口病发病趋势SARIMA模型预测[J].中国公共卫生,2016,32(1):48-52.
[6]刘辉,马殿梅,刘晓坤,等.应用ARIMA-BPNN组合模型预测手足口病发病率[J].现代预防医学,2016,43(16):2885-2888.
[7]赵晶,郭晓雷,吴炳义,等. GM(1,1)灰色预测模型和ARIMA模型在拟合山东省心脑血管疾病死亡率中的应用[J].现代预防医学,2016,43(10):1732-1734,1749.
[8] Liu L,Luan RS,Yin F,et al. Predicting the incidence of hand,foot and mouth disease in Sichuan province,China using the ARIMA model[J]. Epidemiology and Infection,2016,144(1):144-151.
[9]孙霞霞,葛锦荣,李巧方,等. ARIMA模型在宁波市北仑区手足口病发病率预测中的应用[J].现代预防医学,2018,45(4):582-586.
[10]潘欢弘,朱蒙曼,刘晓青. ARIMA乘积季节模型在江西省手足口病发病预测中的应用[J].现代预防医学,2018,45(1):1-4,7.
[11] Sato RC. Disease management with ARIMA model in time series[J]. Einstein(Sao Paulo,Brazil),2013,11(1):128-131.
[12] Wang C,Cao K,Zhang Y,et al. Different effects of meteorological factors on hand,foot and mouth disease in various climates:a spatial panel data model analysis[J]. BMC Infectious Diseases,2016,16:233.
[13]时振东,王华义,王加坤.手足口病及聚集发病与气温关系[J].中国公共卫生,2014,30(12):1586-1588.
[14]王永斌,许春杰,尹素凤,等.中国手足口病发病率ARIMA、RBF及ARIMA-RBF组合模型拟合及预测效果比较[J].中国公共卫生,2017,33(5):760-763.
[2] Xu W,Zhang MX,Qin EQ,et al. Molecular characterization of wild type measles virus from adult patients in northern China,2014[J].International Journal of Infectious Diseases,2016,45(6):36-42.
[3]陈伟,赵晓静,张杰,等.河南省手足口病的发病时间特征分析[J].中国卫生统计,2016,33(2):212-214.
[4]张倩,陈超.改进的GM(1,1)模型在衡水市乙肝发病率预测中的应用[J].现代预防医学,2017,44(11):1925-1928,1937.
[5]杨仁东,胡世雄,邓志红,等.湖南省手足口病发病趋势SARIMA模型预测[J].中国公共卫生,2016,32(1):48-52.
[6]刘辉,马殿梅,刘晓坤,等.应用ARIMA-BPNN组合模型预测手足口病发病率[J].现代预防医学,2016,43(16):2885-2888.
[7]赵晶,郭晓雷,吴炳义,等. GM(1,1)灰色预测模型和ARIMA模型在拟合山东省心脑血管疾病死亡率中的应用[J].现代预防医学,2016,43(10):1732-1734,1749.
[8] Liu L,Luan RS,Yin F,et al. Predicting the incidence of hand,foot and mouth disease in Sichuan province,China using the ARIMA model[J]. Epidemiology and Infection,2016,144(1):144-151.
[9]孙霞霞,葛锦荣,李巧方,等. ARIMA模型在宁波市北仑区手足口病发病率预测中的应用[J].现代预防医学,2018,45(4):582-586.
[10]潘欢弘,朱蒙曼,刘晓青. ARIMA乘积季节模型在江西省手足口病发病预测中的应用[J].现代预防医学,2018,45(1):1-4,7.
[11] Sato RC. Disease management with ARIMA model in time series[J]. Einstein(Sao Paulo,Brazil),2013,11(1):128-131.
[12] Wang C,Cao K,Zhang Y,et al. Different effects of meteorological factors on hand,foot and mouth disease in various climates:a spatial panel data model analysis[J]. BMC Infectious Diseases,2016,16:233.
[13]时振东,王华义,王加坤.手足口病及聚集发病与气温关系[J].中国公共卫生,2014,30(12):1586-1588.
[14]王永斌,许春杰,尹素凤,等.中国手足口病发病率ARIMA、RBF及ARIMA-RBF组合模型拟合及预测效果比较[J].中国公共卫生,2017,33(5):760-763.