ARIMA模型预测南宁市细菌性痢疾的发病趋势

Prediction of Epidemiological Tendency of Bacillary Dysentery in Nanning by ARIMA Model

目的运用自回归移动平均(ARIMA)模型对南宁市2011-2013年的细菌性痢疾发病趋势进行预测,探索细菌性痢疾的流行规律。方法以南宁市2004-2010年的细菌性痢疾月发病率数据为基础,用SPSS13.0建立ARIMA预测模型,以2004-2010年的数据验证模型的预测精度。结果 ARIMA(1,0,1)(0,0,0)12模型可以用于拟合南宁市细菌性痢疾的发病情况。南宁市2011-2013年细菌性痢疾的发病与2010年相比,没有明显的变化,发病呈平稳流行的趋势。结论ARIMA模型可用于细菌性痢疾发病趋势的短期预测。

Objective To predict the epidemiological tendency of bacillary dysentery from 2011 to 2013 by autoregressive integrated moving average（ARIMA） model.Methods Data of bacillary dysentery in Nanning from 2004 to 2010 was used to set up ARIMA prediction model by SPSS13.0.And data from 2004 to 2010 was used to validate model and predict precision.Results ARIMA（1,0,1）（0,0,0）12 model could be used to predict epidemiological tendency of bacillary dysentery in Nanning.Incidence of bacillary dysentery from 2011 to 2013 in Nanning was steady compared with the situation in 2010.Conclusion ARIMA model can be used to make a short-term prediction of bacillary dysentery.