时序预测方法应用探讨——成都市东城区传染病死亡率预测

THE APPLICATION OF TIME SERIES ANALYSIS: A PREDICTION OF THE MORTALITY OF INFECTIOUS DISEASES IN THE EAST DISTRICT OF CHENGDU

本文对成都市东城区1974～1989年传染病死因资料,采用曲线拟合、指数平滑法和B—J模型三类时间序列预测方法对历年传染病总死亡率和位居传染病死因首位的肺结核死亡率进行了拟合,建立了最优预测模型,并进行了预测,为今后防疫工作的决策提供一种有效的预测方法。

The mortality trend of infectious diseases from 1974 to 1989 in the east district of Chengdu was sutdied by curve fitting, exponential smoothing and B—J modeling respectively. The total mortality of infectious diseases and the mortality of lung tuberclosis, the leading death cause of infectious diseases in the period were fitted with those three time senes methods. The best predictive model was established. After applying the modes, we found that there was a strong relationship between the predictive values and the actual data. These results suggested that time series analysis is a valuable tool for determining the basis for controlling infectious diseases.