重庆市HIV感染和艾滋病新发报告总数预测模型的拟合和比较分析

Prediction of new cases of HIV infection and AIDS in Chongqing:data fitting and comparative analysis of 3 prediction models

目的探讨适合于重庆市人类免疫缺陷病毒(human immunodeficiency virus,HIV)感染和艾滋病疫情特点的预测方法,并对近2年的发展变化进行短期预测并验证其结果。方法采用ARIMA模型、指数平滑法和趋势外推法3种常见的时间序列模型对2006-2015年重庆市新报告的HIV/AIDS病例总数进行拟合,对2016和2017年的新报告HIV/AIDS病例数进行预测并验证,构建并比较适合于重庆市艾滋病疫情的模型和预测方法。结果根据模型识别和诊断结果,分别构建了3种时间序列模型用于拟合和预测重庆市HIV/AIDS新发报告总数,分别是ARIMA(0,1,1)(0,1,1)模型、简单季节性指数平滑模型和二次曲线模型。3种模型对流行趋势拟合的标准误差为62. 34、45. 51和227. 29,对近2年新发总数预测的平均相对误差分别为4. 51%、2. 14%和12. 74%。结论 3种预测模型对重庆市HIV/AIDS流行趋势的拟合效果和预测结果不同,简单季节性指数平滑法能更准确地拟合和预测重庆市HIV/AIDS新发总人数。

Objective To establish prediction models for the total number of newly reported cases of HIV infection and AIDS in Chongqing and assess the performance of these models using the data in the recent2 years.Methods The ARIMA model,exponential smoothing method and trend extrapolation method were used to fit the data of the total number of newly reported cases of HIV infection and AIDS in Chongqing collected from 2006 to 2015.The number of newly reported cases in 2016 and 2017 were used to verify the performance of the 3 prediction models to establish the optimal model for epidemic prediction of AIDS in Chongqing.Results Based on the results of model identification and diagnosis,3 time series models,namely ARIMA( 0,1,1)( 0,1,1) model,simple seasonal exponential smoothing model and quadratic curve model,were constructed to fit and predict the total number of new HIV/AIDS cases in Chongqing.The standard errors of the 3 models for fitting the epidemic trends were 62.30,45.51 and 227.29,respectively,and their average relative errors for predicting the total number of new cases in the next two years were4.51%,2.14% and 12.74%,respectively.Conclusion The 3 prediction models have different fitting effects and performance in predicting HIV/AIDS epidemic in Chongqing.Among them,the simple seasonal exponential smoothing method can more accurately fit and predict the total number of new HIV/AIDS cases in Chongqing.