【摘要】 目的探讨时间序列自回归移动平均模型(ARIMA)预测肺癌发病率的可行性。方法收集并整理武汉市江岸区2007～2013年肺癌发病资料,应用SPSS 17.0软件对肺癌逐月发病率进行ARIMA建模拟合;利用建立的预测模型对2013年肺癌月发病率进行预测。结果 ARIMA(3,1,0)模型能够较好地拟合及预测肺癌月发病率值,所有实际值均落在拟合值及预测值的95%置信区间内。结论 ARIMA模型可较好地模拟肺癌发病率在时间序列上的变动趋势,是一种短期预测精度较高的预测模型。更多还原

【Abstract】 Objective To investigate the feasibility of applying time series autoregressive integrated moving average(ARIMA) model in predicting the incidence of lung cancer.Methods Collecting the incidence statistical data from 2007 to 2012 in Jiangan District.SPSS17.0 was used to construct the ARIMA model to predict the monthly incidence of lung cancer in 2013.The incidence from ARIMA model was compared with actual incidence so as to evaluate the model’s validity.Results ARIMA(3,1,0) model could better fit and predict the value of monthly incidence of lung cancer.All real values were within 95% confidence intervals of fitted values and predicted values.Conclusion ARIMA model can well simulate the trend of lung cancer incidence in time series,which is a short-term prediction model with high accuracy.更多还原