应用时间序列模型预测湖北省血吸虫病流行趋势

Prediction of epidemic tendency of schistosomiasis with time-series model in Hubei Province

目的研究湖北省血吸虫病流行的变化趋势,为血吸虫病监测预警提供理论依据。方法运用时间序列ARIMA模型对1987-2013年湖北省居民血吸虫病感染率进行拟合,并预测感染率的短期变化趋势。结果居民血吸虫病感染率的实际值均处于ARIMA模型预测值的95%可信区间内。预测结果显示未来5年湖北省居民血吸虫病感染率仍将继续降低,但下降幅度不大。结论时间序列ARIMA模型预测精度较好,可用于对血吸虫病感染率进行短期预测分析。

Objective,retical basis for surveillance and forecasting of schistosomiasis. Methods The time-series auto regression integrated moving average（ARIMA）model was applied to fit the infection rate of residents of Hubei Province from 1987 to 2013,and to predict the shot-term trend of infection rate. Results The actual values of infection rate of residents were all in the 95% confidence internals of value predicted by the ARIMA model. The prediction showed that the infection rate of residents of Hubei Province would continue to decrease slowly. Conclusion The time-series ARIMA model has good prediction accuracy,and could be used for the short-term forecasting of schistosomiasis.