GM(1,1)灰色模型和ARIMA模型在上海市手足口病发病率预测应用中的比较研究

目的分别采用求和自回归滑动平均模型(autoregressive integrated moving average,ARIMA)和灰色模型GM(1,1)(grey model,GM(1,1))对上海市手足口病的发病率进行预测,并比较两者的预测效果。方法采用2005-2008年上海市手足口病的月发病率和年发病率分别建立ARIMA模型和GM(1,1)模型,并使用2009年的实际年发病率验证两种模型拟合和预测效果。选取相对误差最小的模型预测2011-2012年的发病率。结果 针对手足口病发病率建立的GM(1,1)模型和ARIMA模型,2005-2008年的拟合平均误差率分别为11.06%和10.54%;对2009年进行预测,预测值与实测值的相对误差分别为69.30%和6.51%。采用ARIMA模型对2011年和2012年的年发病率预测为255.32/10万和294.59/10万。结论 ARIMA模型对上海市手足口病的预测效果要优于灰色模型GM(1,1),对解决时间序列类型的发病率等资料有很好的实用价值,预测结果对该病的防治具有科学意义。

Comparison of GM(1,1) gray model and ARIMA model in forecasting the incidence of hand-foot-mouth disease in Shanghai

Objective Integrated autoregressive moving average model(autoregressive integrated moving average,ARIMA) and grey model GM(1,1)(grey model,GM(1,1)) are used to predict the incidence rate of Hand-foot-mouth disease in Shanghai respectively and the predictability of both models are assessed.Methods The ARIMA model and GM(1,1) model were developed based on the monthly and yearly incidence of Hand-foot-mouth disease in Shanghai from 2005 to 2008.Both models were tested by the data in 2009.Relative error was use...