A multivariate tobit analysis of highway accident-injury-severity rates

Panagiotis Ch Anastasopoulos, Venky N. Shankar, John E. Haddock, Fred L. Mannering

Research output: Contribution to journalArticle

102 Citations (Scopus)

Abstract

Relatively recent research has illustrated the potential that tobit regression has in studying factors that affect vehicle accident rates (accidents per distance traveled) on specific roadway segments. Tobit regression has been used because accident rates on specific roadway segments are continuous data that are left-censored at zero (they are censored because accidents may not be observed on all roadway segments during the period over which data are collected). This censoring may arise from a number of sources, one of which being the possibility that less severe crashes may be under-reported and thus may be less likely to appear in crash databases. Traditional tobit-regression analyses have dealt with the overall accident rate (all crashes regardless of injury severity), so the issue of censoring by the severity of crashes has not been addressed. However, a tobit-regression approach that considers accident rates by injury-severity level, such as the rate of no-injury, possible injury and injury accidents per distance traveled (as opposed to all accidents regardless of injury-severity), can potentially provide new insights, and address the possibility that censoring may vary by crash-injury severity. Using five-year data from highways in Washington State, this paper estimates a multivariate tobit model of accident-injury-severity rates that addresses the possibility of differential censoring across injury-severity levels, while also accounting for the possible contemporaneous error correlation resulting from commonly shared unobserved characteristics across roadway segments. The empirical results show that the multivariate tobit model outperforms its univariate counterpart, is practically equivalent to the multivariate negative binomial model, and has the potential to provide a fuller understanding of the factors determining accident-injury-severity rates on specific roadway segments.

Original languageEnglish (US)
Pages (from-to)110-119
Number of pages10
JournalAccident Analysis and Prevention
Volume45
DOIs
StatePublished - Mar 1 2012

Fingerprint

Highway accidents
multivariate analysis
Accidents
accident
Multivariate Analysis
Wounds and Injuries
regression
Statistical Models
Regression Analysis
Databases

All Science Journal Classification (ASJC) codes

  • Human Factors and Ergonomics
  • Safety, Risk, Reliability and Quality
  • Public Health, Environmental and Occupational Health

Cite this

Anastasopoulos, P. C., Shankar, V. N., Haddock, J. E., & Mannering, F. L. (2012). A multivariate tobit analysis of highway accident-injury-severity rates. Accident Analysis and Prevention, 45, 110-119. https://doi.org/10.1016/j.aap.2011.11.006
Anastasopoulos, Panagiotis Ch ; Shankar, Venky N. ; Haddock, John E. ; Mannering, Fred L. / A multivariate tobit analysis of highway accident-injury-severity rates. In: Accident Analysis and Prevention. 2012 ; Vol. 45. pp. 110-119.
@article{a187e32ec90d4746b6efd8d00432cd6d,
title = "A multivariate tobit analysis of highway accident-injury-severity rates",
abstract = "Relatively recent research has illustrated the potential that tobit regression has in studying factors that affect vehicle accident rates (accidents per distance traveled) on specific roadway segments. Tobit regression has been used because accident rates on specific roadway segments are continuous data that are left-censored at zero (they are censored because accidents may not be observed on all roadway segments during the period over which data are collected). This censoring may arise from a number of sources, one of which being the possibility that less severe crashes may be under-reported and thus may be less likely to appear in crash databases. Traditional tobit-regression analyses have dealt with the overall accident rate (all crashes regardless of injury severity), so the issue of censoring by the severity of crashes has not been addressed. However, a tobit-regression approach that considers accident rates by injury-severity level, such as the rate of no-injury, possible injury and injury accidents per distance traveled (as opposed to all accidents regardless of injury-severity), can potentially provide new insights, and address the possibility that censoring may vary by crash-injury severity. Using five-year data from highways in Washington State, this paper estimates a multivariate tobit model of accident-injury-severity rates that addresses the possibility of differential censoring across injury-severity levels, while also accounting for the possible contemporaneous error correlation resulting from commonly shared unobserved characteristics across roadway segments. The empirical results show that the multivariate tobit model outperforms its univariate counterpart, is practically equivalent to the multivariate negative binomial model, and has the potential to provide a fuller understanding of the factors determining accident-injury-severity rates on specific roadway segments.",
author = "Anastasopoulos, {Panagiotis Ch} and Shankar, {Venky N.} and Haddock, {John E.} and Mannering, {Fred L.}",
year = "2012",
month = "3",
day = "1",
doi = "10.1016/j.aap.2011.11.006",
language = "English (US)",
volume = "45",
pages = "110--119",
journal = "Accident Analysis and Prevention",
issn = "0001-4575",
publisher = "Elsevier Limited",

}

A multivariate tobit analysis of highway accident-injury-severity rates. / Anastasopoulos, Panagiotis Ch; Shankar, Venky N.; Haddock, John E.; Mannering, Fred L.

In: Accident Analysis and Prevention, Vol. 45, 01.03.2012, p. 110-119.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A multivariate tobit analysis of highway accident-injury-severity rates

AU - Anastasopoulos, Panagiotis Ch

AU - Shankar, Venky N.

AU - Haddock, John E.

AU - Mannering, Fred L.

PY - 2012/3/1

Y1 - 2012/3/1

N2 - Relatively recent research has illustrated the potential that tobit regression has in studying factors that affect vehicle accident rates (accidents per distance traveled) on specific roadway segments. Tobit regression has been used because accident rates on specific roadway segments are continuous data that are left-censored at zero (they are censored because accidents may not be observed on all roadway segments during the period over which data are collected). This censoring may arise from a number of sources, one of which being the possibility that less severe crashes may be under-reported and thus may be less likely to appear in crash databases. Traditional tobit-regression analyses have dealt with the overall accident rate (all crashes regardless of injury severity), so the issue of censoring by the severity of crashes has not been addressed. However, a tobit-regression approach that considers accident rates by injury-severity level, such as the rate of no-injury, possible injury and injury accidents per distance traveled (as opposed to all accidents regardless of injury-severity), can potentially provide new insights, and address the possibility that censoring may vary by crash-injury severity. Using five-year data from highways in Washington State, this paper estimates a multivariate tobit model of accident-injury-severity rates that addresses the possibility of differential censoring across injury-severity levels, while also accounting for the possible contemporaneous error correlation resulting from commonly shared unobserved characteristics across roadway segments. The empirical results show that the multivariate tobit model outperforms its univariate counterpart, is practically equivalent to the multivariate negative binomial model, and has the potential to provide a fuller understanding of the factors determining accident-injury-severity rates on specific roadway segments.

AB - Relatively recent research has illustrated the potential that tobit regression has in studying factors that affect vehicle accident rates (accidents per distance traveled) on specific roadway segments. Tobit regression has been used because accident rates on specific roadway segments are continuous data that are left-censored at zero (they are censored because accidents may not be observed on all roadway segments during the period over which data are collected). This censoring may arise from a number of sources, one of which being the possibility that less severe crashes may be under-reported and thus may be less likely to appear in crash databases. Traditional tobit-regression analyses have dealt with the overall accident rate (all crashes regardless of injury severity), so the issue of censoring by the severity of crashes has not been addressed. However, a tobit-regression approach that considers accident rates by injury-severity level, such as the rate of no-injury, possible injury and injury accidents per distance traveled (as opposed to all accidents regardless of injury-severity), can potentially provide new insights, and address the possibility that censoring may vary by crash-injury severity. Using five-year data from highways in Washington State, this paper estimates a multivariate tobit model of accident-injury-severity rates that addresses the possibility of differential censoring across injury-severity levels, while also accounting for the possible contemporaneous error correlation resulting from commonly shared unobserved characteristics across roadway segments. The empirical results show that the multivariate tobit model outperforms its univariate counterpart, is practically equivalent to the multivariate negative binomial model, and has the potential to provide a fuller understanding of the factors determining accident-injury-severity rates on specific roadway segments.

UR - http://www.scopus.com/inward/record.url?scp=84455178847&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84455178847&partnerID=8YFLogxK

U2 - 10.1016/j.aap.2011.11.006

DO - 10.1016/j.aap.2011.11.006

M3 - Article

C2 - 22269492

AN - SCOPUS:84455178847

VL - 45

SP - 110

EP - 119

JO - Accident Analysis and Prevention

JF - Accident Analysis and Prevention

SN - 0001-4575

ER -