基于改进诱导有序加权调和平均(IOWHA)算子的传染病组合预测模型研究

Modified IOWHA operator-based study on integrated prediction model for infectious diseases

针对单一模型预测精度不高的问题,提出了一种新的诱导有序加权调和平均(IOWHA)算子和Theil不等系数相结合的传染病组合预测模型,包括ARIMA季节模型和灰色模型,并用其先分别对传染病的发病率进行预测,然后用Theil不等系数与IOWHA算子相结合的方法对预测结果进行组合并求解权重系数。实例应用表明该组合模型的预测精度高于单项模型的预测精度,发挥了各个模型的优势,也说明了组合算法的有效性。

A novel integrated prediction model for infectious diseases combining induction order-weighted harmonic operator and Theil unequal coefficient was proposed in order to solve the poor prediction accuracy of a single model. The novel integrated prediction model is consisted of ARIMA seasonal model and gray model. The incidence of infectious diseases was predicted using ARIMA seasonal model and gray model respectively, which was integrated with the weighted coefficient calculated using the combined Theil unequal coefficient and IOWHA operator. Application of this model in predicting the incidence of infectious diseases showed that its prediction accuracy is higher than that of a single model, and can thus bring the advantages and efficiency of each model into full play.