应用ARIMA模型预测福建省戊型肝炎疫情

摘要

目的 建立福建省戊型肝炎(戊肝)分月发病数预测预警的ARIMA时间序列模型.方法 利用SAS 9.0软件的PROC ARIMA综合软件包对《疾病监测信息报告管理系统》收集的福建省2004-2010年戊肝分月发病数序列进行ARI-MA模型的建模与分析.结果 福建省2004-2010年戊肝分月发病数序列含有以年为周期的季节效应,经12步差分后为平稳非白噪声序列,拟合的相对最优模型为ARIMA(0,0,0)×(0,1,1)12.结论 拟合戊肝的相对最优ARIMA模型进行预测和预警,具有实际应用价值.%In order to construct the ARIMA model for forecasting and warning the monthly cases of hepatitis E in Fujian Province, the time series of monthly cases of hepatitis E from 2004 to 2010 in Fujian Province collected by Disease Information Reporting System were analyzed by the ARIMA of SAS 9. 0. Rersult indicated the time series had an annual seasonal effect, and became stable. And non-white noise series after 12-lag differences was taken. The ARIMA(0,0,0) (0 ,1,1 )12 model was the relative by optimum fitting model. Therefore, fitting a relatively optimum model for hepatitis E is practical for forecasting and warning.