作者: Tom Brijs , Dimitris Karlis , Geert Wets
DOI: 10.1016/J.AAP.2008.01.001
关键词:
摘要: In previous research, significant effects of weather conditions on car crashes have been found. However, most studies use monthly or yearly data and only few are available analyzing the impact daily crash counts. Furthermore, that a level do not explicitly model in time-series context, hereby ignoring temporal serial correlation may be present data. this paper, we introduce an integer autoregressive for modelling count with time interdependencies. The is applied to data, metereological traffic exposure from Netherlands aiming at examining risk observed results show several assumptions related effect counts found if accounted model, produce biased results.