江苏省乙型肝炎流行趋势的时间序列分析及预测

目的：利用拟合ARIMA模型对江苏省乙肝发病趋势进行时间序列分析和预测,为制定乙肝防治策略提供科学依据。方法：收集江苏省2000年～2009年乙肝月发病率资料,通过SPSS13.0软件拟合ARIMA模型。结果：最终拟合为ARIMA（0,1,2）（0,1,1）12模型,残差为白噪声序列,预测值与实际值的平均相对误差为8.26%。结论：江苏省乙肝发病具有逐渐下降的长期趋势,说明乙肝防治卓有成效。2000年以来乙肝月发病率时间序列还有明显的周期性波动特征,其原因有待于进一步研究。2010年预测结果提示需调整本省乙肝防治策略。

Time series analyzing and forecasting to epidemic tendency of hepatitis B in Jiangsu province

Objective：To analyze and forecast incidence of hepatitis B in Jiangsu province with a auto regressive integrated moving average（ARIMA）model,which may provide a scientific basis for developing prevention and control strategies on hepatitis B in the future.Methods：Incidence of hepatitis B was collected monthly from 2000 to 2009 and a model（ARIMA）was fit with SPSS 13.0 software.Rerults：The model ARIMA（0,1,2）（0,1,1）12 was established finally and the residual sequence was a white noise sequence.The relative error in average was 8.26% between the forecasting value and the real value.Conclusion：The incidence of hepatitis B had a long-term descending trend in the last 10 years in Jiangsu province,which declared that the prevention and control on hepatitis B had achieved remarkable success in this area.The incidence sequence of hepatitis B has fluctuated periodically since 2000,and the reason for this is unclear.The predicted values in 2010 suggested that the prevention and control strategies on hepatitis B should be explored further.