自回归滑动平均混合模型在辽宁省人间布鲁氏菌病发病预测中的应用

目的 了解辽宁省2006-2015年人间和家畜布鲁氏菌病（布病）疫情情况，建立时间序列模型，探索畜牧业因素对人间疫情影响，为布病预测防控提供依据。方法 收集辽宁省2006-2016年布病月发病数，以发病数建立时间序列构建自回归滑动平均混合（ARIMA）模型，对2016年1-6月进行短期预测，结合实际观测值评价预测效果；收集辽宁省同期羊存栏量和疫情，分析与人间发病存在的关联。结果 构建模型ARIMA（1，1，1）（1，1，1）12，月发病数拟合值与实际观测值时间序列基本一致，预测效果良好，但精确度较低。羊年存栏量与人间布病发病数存在正关联性。结论 ARIMA模型预测辽宁省布病发病情况基本可行，加强人畜布病的监测工作，以揭示人与家畜流行病学关系。

Application of autoregressive integrated moving average model to predict human brucellosis incidence in Liaoning province

Objective To establish an autoregressive integrated moving average (ARIMA) model to predict human brucellosis incidence in Liaoning,evaluate the influence of animal husbandry on human brucellosis and provide evidence for the prevention and control of human brucellosis. Methods The surveillance data of brucellosis from January 2006 to December 2016 in Liaoning were collected for the establishment of ARIMA model to predict and validate the brucellosis incidence from January to June 2016 in Liaoning. The sheep stock data at the end of a year from 2006 to 2016 were collected to analysis its relationship with human brucellosis. Results The ARIMA(1, 1, 1) (1, 1, 1)12 was established,and the predictive incidence was similar to the actual incidence,but this model showed a lower accuracy. There was a positive correlation between sheep stock at the end of the year and the outbreak of human brucellosis. Conclusion It was feasible to establish a ARIMA model for predicting brucellosis incidence in Liaoning. It is important to strengthen human brucellosis surveillance and find out the the epidemiological correlation between human and livestock.