Spatially Lagged Covariate Model with Zero Inflated Conway-Maxwell-Poisson Distribution Model for the Analysis of Pedestrian Injury Counts

Abstract

Spatial dependency is important to recognize because of the mapping of pedestrian injury counts analysis. Road safety has been a major issue in contemporary societies, with road crashes incurring major human and materials costs annually worldwide. South Korea’s pedestrian traffic accident rate is the highest among the Organization for Economic Cooperation and Development (OECD) countries. In this paper, we use spatially lagged covariate model with zero inflated Conway-Maxwell-Poisson distribution model to account for spatial autocorrelation of no. of pedestrian crashes with cars. Alternatively, the Conway-Maxwell-Poisson (CMP) distribution, first proposed by Conway and Maxwell (1962) has the flexibility to handle all levels of dispersion, including underdispersion. We test spatial autocorrelation of pedestrian injury counts at 2474 sites, with several weights matrices using Moran's I statistics, under permutation scheme. Then we fit different 20 models, and finally choose the best model by the AIC and SBC values.