组合模型对肺结核发病趋势的预测研究

发布时间：2018-06-16 16:03

本文选题：结核 + 肺　；参考：《中国全科医学》2014年21期

【摘要】：目的建立肺结核发病率(1/10万)自回归积分移动平均(ARIMA)-灰色模型(GM)组合模型,并将其应用于肺结核发病率的预测,为及早发现疾病发展趋势和及时采取控制对策提供科学依据。方法收集迁安市2004年1月—2012年12月肺结核月发病率资料,应用SPSS 13.0软件对肺结核逐月发病率进行ARIMA建模拟合;然后用GM(1,1)模型对其带阈值的残差序列进行修正并构造出组合模型,利用此模型对迁安市2013年肺结核逐月发病率进行预测。结果 ARIMA(0,1,1)(0,1,1)12模型较好地拟合了肺结核的发病情况,模型的所有参数都通过统计学检验;用一阈值为4的GM(1,1)模型对其残差序列进行修正,预测模型通过了精度检验(C=0.573,P=0.805),模型拟合精度为基本合格,ARIMA-GM组合模型的平均绝对误差(MAE)、平均绝对百分比误差(MAPE)都比单个模型小,利用组合模型对2013年肺结核发病率预测。结论 ARIMA-GM组合模型能较好地拟合迁安市肺结核发病情况,且该方法比ARIMA季节乘积模型预测具有更高的精度。预测结果能够对肺结核的早期预测预警模型的建立提供借鉴,从而有针对性地采取相应的控制措施。
[Abstract]:Objective to establish an autoregressive integral moving average (ARIMAI-GM) model and apply it to predict the incidence of pulmonary tuberculosis, and to provide scientific basis for early detection of disease development trend and timely control measures. Methods the data of monthly incidence of pulmonary tuberculosis in Qianan City from January 2004 to December 2012 were collected and ARIMA model was used to fit the monthly incidence of pulmonary tuberculosis with SPSS 13.0 software. Then the residual sequence with threshold was modified by using GMM1 / 1) model and a combined model was constructed to predict the monthly incidence of pulmonary tuberculosis in Qianan city in 2013. Results the incidence of pulmonary tuberculosis was well fitted by Arima model, and all the parameters of the model were statistically tested, and the residual sequence was modified by a GM1 1) model with a threshold of 4. The prediction model has passed the accuracy test of 0.573Pu 0.8050.The fitting accuracy of the model is basically qualified. The average absolute error of ARIMA-GM combination model is less than that of single model, and the average absolute percentage error (MAPE) of the model is smaller than that of single model. The combined model is used to predict the incidence of pulmonary tuberculosis in 2013. Conclusion the ARIMA-GM combined model can fit the incidence of pulmonary tuberculosis in Qianan City, and this method is more accurate than the Arima seasonal product model. The prediction results can provide reference for the establishment of early prediction and early warning model of pulmonary tuberculosis and take corresponding control measures.
【作者单位】：河北联合大学公共卫生学院;河北联合大学病理系;河北联合大学附属医院;
【基金】：唐山市科学技术发展课题(131302156) 河北联合大学青年科学研究基金资助项目(z201225)
【分类号】：R521

【相似文献】