论文部分内容阅读
目的比较时间序列季节分解法和差分自回归滑动平均(ARIMA)法预测肺结核发病趋势的效果,为肺结核预测预警提供科学依据。方法对新疆乌鲁木齐市2005年1月至2014年12月肺结核月发病率时间序列分别构建季节分解拟合模型和ARIMA拟合模型,对2015年各月发病率分别进行预测并与实际发病率进行比较。结果乌鲁木齐市肺结核流行表现出春季高发的年度周期性。应用季节分解法构建的拟合模型中,线性模型和三次曲线模型对2015年各月发病率预测结果的平均绝对百分误差(MAPE)分别为18.75%和92.25%,线性模型预测值整体上低于实际值,三次曲线模型预测值整体上高于实际值;应用ARIMA方法构建的拟合模型为ARIMA(2,1,1)(1,1,0)_(12),对2015年各月发病率预测结果的MAPE为9.46%,整体上预测值和实际值无明显差异。结论ARIMA法较季节分解法对乌鲁木齐市肺结核发病率的预测效果更佳。
Objective To compare the effect of time series seasonal decomposition method and differential autoregressive moving average (ARIMA) method in predicting the incidence of pulmonary tuberculosis and provide a scientific basis for predicting and predicting pulmonary tuberculosis. Methods The seasonal decomposition fitting model and ARIMA fitting model were constructed respectively for the monthly incidence of pulmonary tuberculosis from January 2005 to December 2014 in Urumqi, Xinjiang. The morbidity rates of each month in 2015 were respectively predicted and compared with the actual morbidity . Results The prevalence of tuberculosis in Urumqi showed the annual periodic high incidence in spring. The average absolute percent error (MAPE) of the linear model and the cubic curve model for predicting morbidity in each month in 2015 was 18.75% and 92.25%, respectively. The predicted value of the linear model as a whole was low For the actual value, the predicted value of the cubic curve model is higher than the actual value as a whole. The fitting model constructed by ARIMA method is ARIMA (2,1,1) (1,1,0) _ (12) The MAPE of the morbidity prediction result was 9.46%. There was no significant difference between the predicted value and the actual value as a whole. Conclusion The ARIMA method is better than the seasonal decomposition method in predicting the incidence of tuberculosis in Urumqi.