论文部分内容阅读
目的探讨应用时间序列差分自回归移动平均模型(ARIMA)预测北京市东城区艾滋病病毒感染者/艾滋病患者(HIV/AIDS)发病率的可行性。方法应用SAS 9.3软件对北京市东城区2005 2014年HIV/AIDS月发病率构建ARIMA模型,用得到的模型对2015年HIV/AIDS月发病率进行预测,评价模型的拟合和预测效果。结果 ARIMA(0,1,1)×(0,1,1)12模型拟合效果较好,但仍有优化的空间,月发病率的实际值均在预测值的95%可信区间内,预测值与实际值拟合趋势一致。结论 ARIMA模型可以对北京市东城区HIV/AIDS月发病率进行预测,为艾滋病疫情预测提供依据。
Objective To explore the feasibility of using ARIMA to predict the incidence of HIV / AIDS in Dongcheng District of Beijing. Methods SAS 9.3 software was used to construct ARIMA model of monthly incidence of HIV / AIDS in Dongcheng District, Beijing in 2014, and the monthly incidence of HIV / AIDS in 2015 was estimated to evaluate the fitting and prediction effects of the model. Results The ARIMA (0,1,1) × (0,1,1) 12 model fitted well, but there was still space for optimization. The actual incidence of monthly incidence was within 95% confidence interval of the predicted value. The predicted and actual values fit the same trend. Conclusion The ARIMA model can predict the monthly incidence of HIV / AIDS in Dongcheng District of Beijing and provide the basis for the prediction of AIDS epidemic.