Repository logo
 

Refined-scale crash data analysis using multi-level regression models

Simple item page
dc.contributor.authorMa, Xiaoxiang, author
dc.contributor.authorChen, Suren, advisor
dc.contributor.authorArabi, Mazdak, committee member
dc.contributor.authorGrigg, Neil S., committee member
dc.contributor.authorWang, Haonan, committee member
dc.date.accessioned2017-01-04T22:59:17Z
dc.date.available2017-01-04T22:59:17Z
dc.date.issued2016
dc.description.abstractRoad traffic safety has long been a major public health problem for the general public and government agencies. Nevertheless, road traffic crashes continue to bring immeasurable pain and suffering to the society, as well as high financial expenses associated with medical bills and lost productivity. After identifying some key research gaps related to the existing crash modeling such as lack of insightful modeling of crash rates, time-varying explanatory variables, serial correlation, unobserved heterogeneity and multiple dependent variables, the objective of this dissertation is to narrow these gaps by systematically developing advanced multilevel models for traffic safety modeling. It is expected that series of new crash models developed in this dissertation not only contribute to the state-of-the-art crash modeling, but also add to the knowledge toward developing proactive traffic management strategy. The dissertation has eight chapters: Chapter one provides some background information and literature review. Chapter two presents crash rate analysis with data in refined scales to quantify the relation between crash rate and time-varying variables along with other contributing factors. In Chapter three, the unobserved heterogeneity issue on mountainous highways crash rates is examined by developing an advanced random parameter tobit model with panel data in refined temporal scale. Chapter four proposes a correlated random parameter marginalized two-part model as an alternative to study the relationship between crash rate and its contributing factors. Chapter five examines the differences of contributing factors towards injury severity on mountainous (MN) and non-mountainous (NM) highway crashes using mixed logit models. Chapter six studies the effects of weather and traffic characteristics on single-vehicle and multi-vehicle crashes jointly by proposing a multivariate count data model which addresses unobserved heterogeneity across multiple dependent variables. In Chapter seven, a framework of Bayesian multivariate space-time model that can address spatial correlation/heterogeneity, temporal correlation/heterogeneity, and the correlation between different injury severities is introduced. Chapter eight concludes this dissertation by summarizing major findings and sharing some observations in terms of future research.
dc.format.mediumborn digital
dc.format.mediumdoctoral dissertations
dc.identifierMa_colostate_0053A_13956.pdf
dc.identifier.urihttp://hdl.handle.net/10217/178912
dc.languageEnglish
dc.language.isoeng
dc.publisherColorado State University. Libraries
dc.relation.ispartof2000-2019
dc.rightsCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.
dc.subjectmultilevel
dc.subjecttraffic safety
dc.subjectrefined-scale
dc.subjectcrash data analysis
dc.titleRefined-scale crash data analysis using multi-level regression models
dc.typeText
dcterms.rights.dplaThis Item is protected by copyright and/or related rights (https://rightsstatements.org/vocab/InC/1.0/). You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).
thesis.degree.disciplineCivil and Environmental Engineering
thesis.degree.grantorColorado State University
thesis.degree.levelDoctoral
thesis.degree.nameDoctor of Philosophy (Ph.D.)

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Ma_colostate_0053A_13956.pdf
Size:
992.21 KB
Format:
Adobe Portable Document Format
Download

Collections

2000-2019
Theses and Dissertations