|
|||||
|
|
| 基于GRNN的组合预测模型在传染病发病率预测中的应用 | |
| 引用本文： | 叶晓军,沈毅,任茹香,范伟忠. 基于GRNN的组合预测模型在传染病发病率预测中的应用[J]. 浙江预防医学, 2012, 24(1): 8-13 |
| 作者姓名： | 叶晓军 沈毅 任茹香 范伟忠 |
| 作者单位： | 1. 义乌市疾病预防控制中心,浙江 义乌,322000 2. 浙江大学公卫学院 3. 浙江大学公卫学院;绍兴市卫生监督所 |
| 基金项目： | 浙江省科技厅赞助项目（2007C23001） |
| 摘 要： | 目的研究基于GRNN的组合预测模型拟合传染病发病率的优越性和不足。方法以浙中某市1998—2008年的肺结核发病率为研究资料,分别构建了灰色模型和ARIMA模型,以这两种模型为基础构建了基于GRNN的组合预测模型。结果残差修正GM(1,1)模型、ARIMA(1,0,1)*(1,1,0)12模型、基于GRNN的组合预测模型的MSE,MAE,MAPE和MER分别为37.451,5.692,53.69%,48.51%;18.509,3.761,35.13%,32.05%;9.961,2.571,25.6%,21.9%。结论基于GRNN的组合预测模型的预测精度优于两种单项模型。 |
| 关 键 词： | 灰色模型 ARIMA模型 基于GRNN的组合预测模型 发病率预测 |
Application of a Combination Forecasting Model Based on GRNN for Incidence of Pulmonary Tuberculosis |
|
| Affiliation: | YE Xiao -jun, SHEN Yi, REN Ru -xiang, et al. Yiwu Municipal Center for Disease Control and Prevention, Yiwu, Zhefiang, 322000, China. |
| Abstract: | Objective To establish a combination forecasting model based on GRNN which could be used to predict the incidence rate of the infection diseases and evaluate its advantages and weakness. Methods Grey model and ARIMA model were established with the pulmonary tuberculosis incidence of the city in middle of Zhejiang province from 1998 to 2008. GRNN combination forecasting model was established based on the Grey and ARIMA model. Results MSE, MAE, MAPE and MER of GM （ 1, 1 ） model residual error correction, ARIMA （ 1, 0, 1 ） ＊ （ 1, 1, 0） 12, GRNN combination forecasting model were 37. 451,5. 692, 53. 69%, 48. 51% ; 18. 509, 3. 761, 35.13%, 32. 05 % ; 9. 961, 2. 571, 25.6%, 21.9%, respectively. Conclusion The forecast accuracy of the GRNN combination forecasting model was better than single GM model and single ARIMA model. |
| Keywords: | Grey model ARIMA model GRNN combination forecasting model Incidence forecast |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |