基于ARIMA的新城疫风险预测模型

Risk Prediction Model for Newcastle Disease Based on ARIMA

本研究旨在探讨ARMIA模型在新城疫风险预测方面的可行性,主要以地区为单位预测新城疫月新发生次数,为制定科学的防控措施提供指导。通过分析2009年1月到2012年12月的新城疫月新发生次数的官方数据,建立具有时间序列属性的风险预测模型,并对所建立的模型进行随机性、显著性和拟合度检验,从而确立风险预测模型中的参数,并使用分析预测模型对2013年1月到2013年6月的新城疫月新发生次数进行预测。将模型预测结果与实际新城疫月新发生次数进行对比表明,基于ARIMA模型的新城疫风险预测误差较小,效果比较显著。

The aim of this paper is to study the feasibility of ARIMA model in Newcastle disease risk forecasting, it mainly focuses on forecasting the number of monthly new outbreaks, which can be used to reasonably guide the policy--making of prevention and control the disease. By analyzing the official data from Jan. 2009 to Dec. 2012, we establish a risk forecasting model with time series properties, and test the ran- domness, significant and fitting of our model to calculate the parameters in the model. Then we use this model to forecast the data from Jan. 2013 to Jun. 2013. Comparing the forecasting result with the actual data, it shows that the error of ARIMA--based risk forecasting model is lower and the forecasting result is remarkable. Under the assumption of no significant changes in environments and time--space factors, the static forecasting of Newcastle disease based on ARIMA model is feasible and has the positive effect in disease control.