成都市人口死亡率时间序列分析与预测

目的采用时间序列分析和预测成都市人口死亡率的动态发展趋势,建立时间序列模型,考察模型的应用效果并做出预测。方法利用时间序列自相关系数和偏相关系数识别模型,采用最小二乘法估计模型参数,用Box-Ljung统计量评价ARIMA模型的拟和度,用平均预测相对误差作为预测效果的评价指标。结果建立乘积ARIAM(0,1,1)(0,1,1)12模型,模型平均绝对百分误差MAPE=8.50%。成都市人口死亡率自2000年逐渐下降,预计序列后2年将继续呈现下降趋势。结论所运用的时间序列分析和预测模型拟合效果较好,可应用于疾病发病和死亡动态变化规律的分析和其未来发展趋势的预测、预报。

Time Series Analysis and Forecast of Mortality Rate in Chengdu

Objective To analyze and forecast the dynamic trend of mortality rate in Chengdu by time series. And to build a time series model and evaluate its predictive effect. Methods ACF and PACF analysis were used to identify the model. The model parameter was estimated by least square method. Statistics of Box-Ljung was used to evaluate the degree of fitness of ARIMA mode1, and the average relative errors of prediction were used as indexes to evaluate the predictive effect. Results The multiple seasonal ARIMA(0,1,1)(0,1,1)12 model showed the average relative errors reached 8.50%. The morality rate descended gradually after 2000, and the forecast showed that the mortality rate would decrease continually in the next 2 years. Conclusion The fitness of time series analysis and forecast model is good,so it can be used to analyze the regularity of disease or death variation and forecast its future tendency.