SARIMA-SVM组合模型在肺结核发病率预测中的应用

Application of SARIMA-SVM Model in Tuberculosis Incidence

利用2004-2015年甘肃省肺结核逐月发病疫情资料,论文构建了SARIMASVM组合模型,并将其应用于肺结核发病率的预测.根据数据特征组合模型的构建采用串联方式,进而利用均方误差(MSE)、平均绝对百分误差(MAPE)等指标评价所建模型的内插拟合效果,以比较单一SARIMA模型与SARIMA-SVM组合模型在肺结核发病率预测中的精度.结果表明:SARIMA-SVM组合预测模型能有效提高模型的预测精度,其预测效果优于单一SARIMA模型,具有较强的实用性与泛化能力.基于此组合模型,论文就甘肃省肺结核发病率做出外推预测,以期为管理与控制甘肃省肺结核疾病提供科学定量依据.

Based on the monthly incidence of tuberculosis in Gansu Province from 2004 to 2015,the paper constructed a SARIMA-SVM combined model and applied it to the prediction of tuberculosis incidence.According to the combination of data characteristics,the model is connected in series,and the indicators such as mean square error(MSE)and mean absolute percentage error(MAPE)are used to evaluate the interpolation fitting efect of the model,to compare the accuracy of single SARIMA model and SARIMA-SVM combined model in predicting the incidence of tuberculosis.The results show that the SARIMA-SVM combined prediction model can effectively improve the prediction accuracy of the model,and its prediction effect is better than that of the single SARIMA model,and has strong practicability and generalization ability.Based on this combined model,the paper makes extrapolated predictions of the incidence of tuberculosis in Gansu Province,with a view to providing scientific quantitative basis for the management and control of tuberculosis in Gansu Province.