论文部分内容阅读
目的通过ARIMA模型结合圆分布法对1999年1月—2006年12月天津市冠心病逐月死亡率数据进行分析,研究其动态变化规律。方法应用圆分布法探讨冠心病死亡率的季节分布。通过模型辨识、参数估计及其检验、模型的拟合度分析等过程,建立ARIMA季节乘积模型(p,d,q)(P,D,Q)S。结果冠心病死亡率以年为周期,一年中1月为高发月份。建立了ARIMA(0,1,0)×(0,1,1)12模型:(1-B)(1-B12)lnxt=0.002+(1-0.421B)(1-0.796B12)εt。结论ARIMA乘积模型结合圆分布法是对冠心病死亡率进行时间序列分析的重要方法,可对今后的冠心病流行趋势及死亡率进行预测。
Objective To analyze the data of monthly mortality of coronary heart disease in Tianjin from January 1999 to December 2006 by ARIMA model combined with circular distribution method and to study its dynamic changes. Methods The circular distribution of coronary heart disease mortality rate of the seasonal distribution. The ARIMA seasonal product model (p, d, q) (P, D, Q) S is established through the process of model identification, parameter estimation and testing, and fitting of the model. Results The mortality rate of coronary heart disease was annual, and January was the high incidence month in January. The ARIMA (0,1,0) × (0,1,1) 12 model is established: (1-B) (1-B12) lnxt = 0.002 + (1-0.421B) (1-0.796B12) εt. Conclusions The ARIMA product model combined with circular distribution method is an important method for time series analysis of coronary heart disease mortality, which can predict the future trend of coronary heart disease and mortality.