我国恶性肿瘤死亡人数比重预测的ARIMA模型

ARIMA Models for Predicting the Proportion of Cancer Deaths in China

目的:利用差分整合移动平均自回归模型(autoregressive integrated moving average,ARIMA)预测我国城市与农村的恶性肿瘤死亡人数占总死亡人数百分比,为恶性肿瘤防护工作与医疗资源合理分配提供参考依据。方法:利用国家统计局的国家数据官网公布的年度数据中,我国城市与农村恶性肿瘤死亡人数占比数据,分别建立ARIMA模型,并通过模型参数分析,预测值与真实值的对比分析,总结模型合理性。结果:通过参数计算与模型验证,最终得到城市ARIMA(2,1,0)和农村ARIMA(2,1,0)模型。结论:通过模型预测验证,模型可以用于我国城市与农村恶性肿瘤死亡率的短期预测与动态分析,具有实际应用价值。

Objective:To predict the percentage of cancer deaths in China’s urban and rural areas by using autoregressive integrated moving average(ARIMA)models,which provides a reference for the prevention of cancer and reasonable allocation of medical resources.Methods:According to the annual data from National Bureau of Statistics,ARIMA model was established to analyze the proportion of cancer deaths in urban and rural areas of China,and the rationality of the model was summarized by analyzing parameters of the model and comparing predicted values with real values.Results:Through parameter calculation and model validation,urban ARIMA(2,1,0)and rural ARIMA(2,1,0)models were finally obtained.Conclusion:ARIMA models can be used for short-term prediction and dynamic analysis of cancer mortality in urban and rural areas of China,which is practically applicable and valuable.