基于ARIMA乘积季节模型的苏州市介水传染病发病预测研究

王建书; 刘强; 覃江纯; 杭惠; 杨海兵

doi:10.13421/j.cnki.hjwsxzz.2017.06.001

基于ARIMA乘积季节模型的苏州市介水传染病发病预测研究

Prediction of Incidence for Water-borne Diseases on a Multiple Seasonal ARIMA Model in Suzhou

摘要

摘要:
目的探讨运用自回归求和移动平均（autoregressive integrated moving average，ARIMA）乘积季节模型对苏州市介水传染病发病率进行预测。
方法利用R软件对苏州市2008年1月—2015年12月的介水传染病发病率数据进行拟合，构建ARIMA乘积季节模型，对苏州市2016年1—6月介水传染病的发病率进行预测。
结果构建了ARIMA（2，1，2）×（0，1，1）₁₂乘积季节模型，模型Ljung—Box检验差异无统计学意义（Q=18.478，P=0.779），模型适用于短期预测，2016年1—6月苏州市常见介水传染病实际发病率均在预测结果95%可信区间内，预测结果相对误差的平均值为-0.024。
结论 ARIMA（2，1，2）×（0，1，1）₁₂季节乘积模型可用于苏州市介水传染病发病率的短期预测。

Abstract:
Objectives To explore the application of a multiple seasonal autoregressive integrated moving average (ARIMA) model in predicting the incidence of water-borne diseases in Suzhou.
Methods A multiple seasonal ARIMA model based on the incidence of water-borne diseases in Suzhou from 2008 to 2015 was established by the R software, an optimal fitted model was then used to predict the incidence of water-borne diseases in Suzhou from January to June in 2016.
Results A multiple seasonal ARIMA (2, l, 2)×(0, l, 1)₁₂ model was established. There was no statistically significant difference in fitting effect(Q=18.478, P=0.779)tested by the Ljung-Box test, and the model was fitted for forecasting a short term incidence rate, all actual values of the incidence of water-borne diseases in Suzhou from January to June in 2016 were in the 95% confidence intervals of predicted values, the mean relative error was -0.024.
Conclusion Multiple seasonal ARIMA (2, l, 2)×(0, 1, 1)₁₂ model could be used to predict the short term incidence rate of water-borne diseases in Suzhou.

HTML全文

参考文献(12)

施引文献

资源附件(0)