基于时间序列模型的兰州市5岁以下儿童死亡预测分析

Prediction and analysis of death of children under 5 years old in Lanzhou based on time series model

目的分析2010-2018年兰州市5岁以下儿童死亡情况,构建时间序列模型预测2019年兰州市5岁以下儿童死亡趋势。方法采用描述流行病学方法综合分析兰州市2010年1月-2018年12月5岁以下儿童死亡情况,利用SPSS 21.0软件构建时间序列分析模型,筛选最佳模型并预测2019年兰州市5岁以下儿童死亡情况。结果兰州市2010-2018年共报告5岁以下儿童死亡病例1650例,男、女报告死亡例数分别为871例和774例,年均死亡率为6.23‰。近几年兰州市5岁以下儿童死亡率总体呈下降趋势;5岁以下儿童死亡以新生儿为主,占65.27%;通过不同模型比较发现简单季节性指数平滑模型为最优模型,该模型较好的拟合了兰州市2010-2018年5岁以下儿童月死亡例数,预测2019年兰州市5岁以下儿童总死亡例数为140例,与2018年的死亡例数相近。结论兰州市5岁以下儿童死亡率总体呈逐年下降趋势,简单季节性指数平滑模型可以较好的反映兰州市5岁以下儿童的死亡趋势并进行短期预测。

Objective To analyze the death trend of children under 5 years old in Lanzhou and establish the time series model to predict the mortality and incidence of children under 5 years old in Lanzhou in 2019.Methods Descriptive epidemiological method was used to analyze the mortality of children under 5 years old in Lanzhou from January 2010 to December 2018.SPSS 21.0 software was used to construct time series analysis model,selecting the best model and predict the mortality of children under 5 years old in Lanzhou in 2019.Results A total of 1650 deaths of children under 5 years old were reported in Lanzhou from 2010 to 2018.The number of deaths reported by boys and girls was 871 and 774 respectively,with an average annual mortality rate of 6.23‰.In recent years,the overall mortality rate of children under 5 years old in Lanzhou had declined.The majority of deaths among children under 5 years old were neonates,accounting for 65.27%.Simple seasonal model was the best model by comparing different models.The model could well fit the monthly death cases of children under 5 years old in Lanzhou from 2010 to 2018.It is predicted that the total number of deaths of children under 5 years old in Lanzhou will be 140 in 2019,which is similar to the number of deaths in 2018.Conclusions The mortality rate of children under 5 years old in Lanzhou is decreasing year by year.Simple seasonal model can better reflect the mortality trend of children under 5 years old in Lanzhou and make short-term prediction.