摘要
目的探讨建立残差自回归模型预测某市艾滋病新发感染率的可行性。方法根据某市40个区县1995-2009年艾滋病感染情况数据资料,建立了二次方程和ARMA(1,0)的残差自回归模型。结果残差自回归模型能较好地拟合当地艾滋病月度新发感染率的变动趋势,且2009-08/12的预测值符合实际值的变动趋势。结论该模型能够较好地模拟当地艾滋病新发感染率在时间序列上的变化,可为该地艾滋病新发感染情况的短期预测提供依据。
Objective To explore the feasibility of Auto-regressive model on forecasting the HIV/AIDS infection rates.Methods Data of HIV/AIDS infection of 40 administrative area of A city from 1995 to 2009 were collected.Quadratic equation and ARMA（1,0） of the Auto-regressive model were established.Results The Auto-regressive model was capable of fitting the AIDS infection rates of new cases with monthly changing trend,and also the actual trend from August 2009 to December 2009.Conclusion The model can exactly fit the incidence of AIDS in time series,which can be used for short-term prediction.
出处
《预防医学情报杂志》
CAS
2013年第3期175-177,共3页
Journal of Preventive Medicine Information
基金
国家自然科学基金项目
项目批准号:81001295
