A copula-based approach to accommodate the dependence among microscopic traffic variables

摘要: Abstract Developing microscopic traffic simulation models requires the knowledge of probability distributions variables. Although previous studies have proposed extensive mathematical for describing variables (e.g., speed, headway, vehicle length, etc.), these usually consider observations to be independent and are investigated separately. As a result, some traditional approaches as inputs process methods may ignore possible dependence among different The objectives this paper investigate structure examine applicability copula approach joint modeling Copulas functions that relate multivariate distribution random their one-dimensional marginal functions. concept copulas has been well recognized in statistics field recently introduced transportation studies. is applied 24-h data collected on IH-35 Austin, Texas. preliminary analysis indicates there exists Moreover, results suggest can adequately accommodate accurately reproduce revealed by observations. Overall, findings provide framework generating multiple simultaneously considering dependence.