GM(1,1)模型在围生儿死亡率预测中的应用

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

目的探讨灰色预测模型在围生儿死亡率预测中的应用,为妇幼保健工作提供科学数据。方法利用深圳市某街道2013—2019年的围生儿死亡率的数据构建GM(1,1)模型,评价拟合效果。结果深圳市某街道常住人口围生儿死亡率GM(1,1)模型为:Y(t+1)=-29.29426e-0.19917t+37.77426;流动人口围生儿死亡率GM(1,1)模型为:Y(t+1)=-110.43433e-0.08068t+119.53433;总人口围生儿死亡率GM(1,1)模型为:Y(t+1)=-41.37463e-0.18837t+50.29463。2018年常住人口、流动人口及总人口的围生儿死亡率预测值分别为:1.6‰、5.28‰及2.29‰。上述三个模型后验差比值C均小于0.35,小误差概率均为P=1。结论建立的模型预测结果与实际情况相近,模型拟合误差较小,GM(1,1)模型对中国围生儿死亡率预测效果较好,呈逐年下降趋势。

Objective To explore the application of gray prediction model in the predicting of perinatal mortality and provide scientific data for maternal and child health care work.Methods Using the data of perinatal mortality in one sub-district in Shenzhen from 2013 to 2019,the GM(1,1)model was constructed and the fitting effect was evaluated.Results The perinatal mortality rate GM(1,1)model for resident population in a sub-district in Shenzhen is:Y(t+1)=-29.29426e-0.19917t+37.77426;The perinatal mortality rate GM(1,1)model for floating population is:Y(t+1)=-110.43433e-0.08068t+119.53433.The perinatal mortality rate GM(1,1)model for the total population is:Y(t+1)=-41.37463e-0.18837t+50.29463.The estimated perinatal mortality rates of the population,the floating population and the total population in 2018 are 1.6‰,5.28‰and 2.29‰,respectively.The posterior difference ratios C of the above three models are less than 0.35,and the probability of small errors is P=1.Conclusions The model predictive results are similar to the actual ones,and the error of the model fitting is small.The GM(1,1)model has a good effect on predicting perinatal mortality in China,showing a declining trend year by year.