Research on the Yangtze River accident casualties using zero-inflated negative binomial regression technique

Abstract

The zero-inflated negative binomial regression model is used to study the influencing factors of casualties in water traffic accidents. The Pearson correlation analysis is used to obtain the influencing factors of casualties in water traffic accidents. The ZINB regression model is established to evaluate the influencing factors. According to the analysis results, influencing factors are sorted in the ascending order of factor influencing value. This method has been applied in casualties in water traffic, which is in the area under administration of Yangtze River MSA (Maritime Safety Administration). And the results can provide the reference for the maritime management to make effective strategy to reduce the casualties in water traffic accidents.