基于季节性(差分整合)自回归移动平均模型的广西乙类传染病发病情况预测

Prediction of the incidence of category B infectious diseases in Guangxi based on seasonal autoregressive integrated moving average model

目的应用季节性(差分整合)自回归移动平均(SARIMA)模型预测广西乙类传染病发病情况。方法将2011年1月至2022年5月广西乙类传染病月报告发病数据作为训练集构建时间序列,拟合和构建SARIMA预测模型;以2022年6月至11月的广西乙类传染病月报告发病数据作为测试集对模型进行测试。结果广西乙类传染病的发病情况呈季节性规律,最优预测模型为SARIMA(3,1,3)(2,0,0)_(12),其预测效果平均相对误差为7.99%,预测发病例数95%CI均包含了实际发病例数。结论SARIMA(3,1,3)(2,0,0)_(12)模型能较好地拟合广西乙类传染病的发病情况,可用于疫情的短期监测。

Objective To predict the incidence of category B infectious diseases in Guangxi with the seasonal autoregressive integrated moving average(SARIMA)model.Method The monthly reported incidence data of category B infectious diseases in Guangxi from January 2011 to May 2022 were used as the training set to construct a time series,and the SARIMA prediction model was fitted and constructed.The monthly reported incidence data of category B infectious diseases in Guangxi from June to November 2022 was used as the test set to test the model.Results The incidence of category B infectious diseases in Guangxi was seasonal,and the optimal prediction model was SARIMA(3,1,3)(2,0,0)_(12),with a mean relative error of the predictive validity of 7.99%,and the 95%CI of the predictive incidence included the actual incidence.Conclusion The SARIMA(3,1,3)(2,0,0)_(12) model can well fit the incidence of category B infectious diseases in Guangxi and can be used for short-term surveillance of the epidemic.