北京地区新生儿先天性甲状腺功能低下症发病率的ARIMA预测模型研究

Study on the prevalence rate of Neonatal Congenital Hypothyroidism of ARIMA Model in Beijing

目的建立北京地区先天性甲状腺功能低下症(CH)发病率的预测模型,为合理制定CH的防治、保健的策略及措施提供科学依据。方法构建并应用自回归求和移动平均(autoragressive integrated moving average,ARIMA)模型预测北京市的CH发病率。结果 ARIMA(0,1,0)模型可以较好的拟合1994-2011年北京市CH发病率的时间变化趋势,并可用于预测未来的CH发病率,提供较准确的CH发病率预测。根据预测值得知,2012年以后发病率可能会有较为明显的上升趋势,未来2年北京市CH平均发病率为42.02/10万。结论 ARIMA(0,1,0)模型可以较好的拟合北京市CH发病率的时间变化趋势,提示2012年以后发病率可能会有较为明显的上升趋势。

Objective： Establishing congenital hypothyroidism＇s predictive model in Beijing, that is to provide scientific support for the establishment of reasonable strategy and measure for the prevention of CH. Methods ： Establishing and applying of autoregressive in- tegrated moving average （ARIMA） model to predict the prevalence rate of CH in Beijing. Results： The result shows ARIMA （0, 1, 0） model can be used to fit the changes of prevalence rate of CH and to forecast the future prevalence rate in Beijing. The predicted prevalence rate of CH in 2012 and 2013 based on the prevalence rate would be 42. 02 per 100 thousand. It is a predicted model of high precision for short - ime forecast. Conclusion： ARIMA （0, 1, 0） model can be used to fit the changes of prevalence rate of CH and to forecast the future prevalencerate in Beijing.