Analysis Of Factors Influencing The Number And Severity Of Traffic Accidents

Traffic accidents not only cause casualties,but also have an impact on economic development and social stability.Based on this,it is of great significance to model and analyze the factors influencing the number and severity of traffic accidents and put forward targeted suggestions.(1)Modeling and analysis of the number of traffic accidents considering spatial effects under the panel dataFirstly,the panel data of 31 provinces(excluding Hong Kong,Macao and Taiwan)in China in the 16 years from 2006 to 2021 are selected for descriptive statistical analysis,and the conclusion that there is a significant spatial positive correlation in the number of traffic accidents in China by drawing the spatiotemporal evolution map,calculating the global Moran index,plotting the local Moran scatter plot,LISA significance map and agglomeration map.Secondly,the influencing factors are selected and a spatial econometric model is established,the fixed effect,random effect and Hausman test are carried out respectively when the spatial panel data model is selected and tested,because the fixed effect and random effect are significant and the Hausman test results reject the null hypothesis,so the random effect is not selected when establishing the model.Finally,the spatial autoregressive model(SAR),spatial error model(SEM)and spatial Duberman model(SDM)are established for comparative analysis,according to the estimation results of the three models,it is found that the spatial Dubin model(SDM)under the time fixation effect is the most significant and the fitting effect is the best.Therefore,the spatial Dubin model(SDM)under the time fixation effect is selected as the final model for analysis.On this basis,in order to further examine the relationship between GDP per capita and the number of traffic accidents,this paper further introduces the panel vector autoregressive model(PVAR)to conduct empirical research on per capita GDP and traffic accidents.The study shows that the per capita GDP,illiterate numbers,number of hospitals,and per capita urban road areas of 31 provinces in China(excluding Hong Kong,Macao and Taiwan)plays a role in promoting the number of traffic accidents,and the number of highway mileage of 10,000 people,the number of public transport vehicles,the number of urban lights per kilometer plays a restraining role.At the same time,there is a dynamic relationship between GDP per capita and the number of traffic accidents.(2)Modeling and analysis of the traffic accident severity based on multiclassification logistic regression model(Multinomial Logit)Firstly,the severity of traffic accidents and index variable data are calibrated.secondly,by using multi-classification logistic regression model(Multinomial Logit)for modeling and analysis,it is concluded that there are no intersections,no stations,and no traffic signs nearby play a role in restraining the severity of traffic accidents,and the increase in rainfall in the weather plays a promoting role,while being in the daytime has a promoting effect on fewer disability accidents and has a restraining effect on fatal accidents.At the same time,the absence of humps nearby has no obvious effect on severity.And from the analysis of the results,it can be seen that relative to the only property damage accidents,the most critical influencing factors in fewer disability accidents,more disability accidents and fatal accidents are whether there is an intersection.Finally,based on the analysis of conclusions,this paper gives some suggestions to reduce the number of traffic accidents and reduce the severity of accidents.