摘要
目的探讨应用自回归滑动平均混合模型(ARIMA)预测肺结核发病率的可行性。方法应用SPSS13.0软件对甘肃省2000年1月～2008年6月肺结核逐月发病率进行ARIMA建模拟合;用2008年7～12月逐月发病率检验模型,并进一步预测2009年分月发病情况。结果模型ARIMA(0,1,1)(1,1,0)12(不含常数项)所有参数都通过统计学检验,残差序列是白噪声,拟合优度相对最好,参数间弱相关(r=0.269),用模型进行预测,其预测值与实际值之间的平均相对误差为0.046,模型预测精度较高。结论自回归滑动平均混合模型可很好地模拟和预测肺结核发病率在时间序列上的变动趋势,将其应用于肺结核发病预测是可行的。
Objective To explore the application of auto regressive integrated moving average（ARIMA）model in prediction of pulmonary tuberculosis incidence.Methods SPSS13.0 software was used to construct the ARIMA model based on the monthly pulmonary tuberculosis incidence of GanSu province from Jan.2000 to Jun.2008.Pulmonary tuberculosis actual incidence of Jul.2008 to Dec.2008 was compared with incidence which predicted by ARIMA model to test model,and predicting the incidence of 2009 further.Results ARIMA（0,1,1）（1,1,0）12（have no constant）model was constructed.Comparing actual value with the predicted value,the average relative error was 0.046,and the model prediction is accuracy.Conclusion The model exactly fitted and predicts the changes and trends of pulmonary tuberculosis incidence on time series,thus,ARIMA model applied to predict pulmonary tuberculosis incidence in GanSu province is feasibility.
出处
《中国卫生统计》
CSCD
北大核心
2010年第5期507-509,共3页
Chinese Journal of Health Statistics
关键词
ARIMA模型
预测
肺结核
ARIMA model
Predict
Pulmonary tuberculosis
