基于改进灰色预测模型的交通事故预测模型研究

Research on traffic accident prediction model based on improved grey prediction model

为了提高交通事故数据预测的准确度,采取GM(1,1)和OSDGM(1,1)等单一模型,对2008-2019年我国交通事故死亡人数数据进行分析。根据GM(1,1)和OSDGM(1,1)模型建立最优加权组合模型,使用Verhulst模型对建立的加权组合模型进行残差修正,并借助灰色模型精度评价指标对预测结果进行检验。预测结果表明,GM(1,1)、OSDGM(1,1)模型和改进的灰色预测模型的预测结果的平均相对误差分别为4.35%、4.30%和1.19%,且改进的灰色模型通过精度指标检验,说明改进灰色预测模型具有较高的精度。

In order to improve the accuracy of traffic accident data prediction, a single model such as GM(1,1) and OSDGM(1,1)is used to analyze and predict the traffic accident fatalities data in China from 2008 to 2019. According to the GM(1, 1) and OSDGM(1, 1) models, the optimal weighted combination model is established. The Verhulst model is used to correct the residuals of the established weighted combination model, and the prediction results are tested with the grey model accuracy evaluation index.The prediction results show that the average relative errors of the prediction results of the GM(1,1), OSDGM(1,1) models and the improved grey prediction model are 4.35%, 4.30% and 1.19%, respectively. The improved grey model has passed the accuracy index test. It shows that the improved grey prediction model has higher accuracy.