GM(1,1)模型在中国围产儿死亡率预测中应用

Application of gray model(1,1) in prediction of perinatal mortality rate in China

目的应用SAS 9.3软件建立灰色GM(1,1)预测模型,对中国围产儿死亡率进行预测,为妇幼保健工作提供参考。方法以中国2005—2013年围产儿死亡率数据为基础,建立GM(1,1)模型并进行预测。结果GM(1,1)预测模型为^X(1)(t+1)=-124.2050e-0.0816598t+134.4750,预测结果显示,我国2014—2016年围产儿死亡率分别为5.067 81‰、4.670 42‰、4.304 19‰;各种模型检验结果显示,模型的预测精度等级较高,可用于外推。结论 GM(1,1)模型对中国围产儿死亡率预测效果较好,呈逐年下降趋势,该预测结果对我国妇幼工作有一定的指导意义。

Objective To establish a gray model （ 1,1 ） to forecast perinatal mortality rate in China with SAS 9. 3 software and to provide references for maternal and child health care. Methods The data on perinatal mortality from 2005 to 2013 in China were collected and a gray model（ 1,1 ）for perinatal mortality prediction was established based on the data. Results The mathematic formula for the model established was X（1） （t＋1） = - 124. 2050e^-0.0816598t ＋ 134. 4750. The predicted perinatal mortality rates in China for the years of 2014,2015, and 2016 are 5. 06781‰, 4. 67042‰, and 4. 30419‰, respectively. The results of verification with various other models showed that the predictive accuracy of the model established is at a high level and the model could be used for extrapolation. Conclusion The prediction on perinatal mortality rate in China based GM （ 1,1 ） model is valid and the prediction is valuable for maternal and child health care in China.