Accident Count Model Based on Multiyear Cross-Sectional Roadway Data with Serial Correlation

Gudmundur F. Ulfarsson, Venkataraman Shankar

Research output: Contribution to journalArticle

48 Citations (Scopus)

Abstract

The use of the negative multinomial model to form a predictive model of median crossover accident frequencies with a multiyear panel of cross-sectional roadway data with a roadway section-specific serial correlation across time was explored. The negative multinomial model specification is compared with previous research, which used the same database but which also used negative binomial and random-effects negative binomial count models. If there is no section-specific correlation in the panel, the negative multinomial model becomes equivalent to the negative binomial. The differences in the estimation results between those models show that such a correlation exists in the data. The results show that the negative multinomial significantly outperforms the negative binomial and the random-effects negative binomial in terms of fit, with a statistically significantly higher likelihood at convergence. The signs of the coefficients were similar hi all models; when the signs differed, the negative multinomial model results were more intuitive. Overall, the analysis supports the use of the negative multinomial count model to estimate median crossover accident frequency models that are based on panel data.

Original languageEnglish (US)
Pages (from-to)193-197
Number of pages5
JournalTransportation Research Record
Issue number1840
DOIs
StatePublished - Jan 1 2003

Fingerprint

Accidents
Specifications

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Mechanical Engineering

Cite this

Ulfarsson, Gudmundur F. ; Shankar, Venkataraman. / Accident Count Model Based on Multiyear Cross-Sectional Roadway Data with Serial Correlation. In: Transportation Research Record. 2003 ; No. 1840. pp. 193-197.
@article{30c08b6c28b74a4c990c116b27b5bab7,
title = "Accident Count Model Based on Multiyear Cross-Sectional Roadway Data with Serial Correlation",
abstract = "The use of the negative multinomial model to form a predictive model of median crossover accident frequencies with a multiyear panel of cross-sectional roadway data with a roadway section-specific serial correlation across time was explored. The negative multinomial model specification is compared with previous research, which used the same database but which also used negative binomial and random-effects negative binomial count models. If there is no section-specific correlation in the panel, the negative multinomial model becomes equivalent to the negative binomial. The differences in the estimation results between those models show that such a correlation exists in the data. The results show that the negative multinomial significantly outperforms the negative binomial and the random-effects negative binomial in terms of fit, with a statistically significantly higher likelihood at convergence. The signs of the coefficients were similar hi all models; when the signs differed, the negative multinomial model results were more intuitive. Overall, the analysis supports the use of the negative multinomial count model to estimate median crossover accident frequency models that are based on panel data.",
author = "Ulfarsson, {Gudmundur F.} and Venkataraman Shankar",
year = "2003",
month = "1",
day = "1",
doi = "10.3141/1840-22",
language = "English (US)",
pages = "193--197",
journal = "Transportation Research Record",
issn = "0361-1981",
publisher = "US National Research Council",
number = "1840",

}

Accident Count Model Based on Multiyear Cross-Sectional Roadway Data with Serial Correlation. / Ulfarsson, Gudmundur F.; Shankar, Venkataraman.

In: Transportation Research Record, No. 1840, 01.01.2003, p. 193-197.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Accident Count Model Based on Multiyear Cross-Sectional Roadway Data with Serial Correlation

AU - Ulfarsson, Gudmundur F.

AU - Shankar, Venkataraman

PY - 2003/1/1

Y1 - 2003/1/1

N2 - The use of the negative multinomial model to form a predictive model of median crossover accident frequencies with a multiyear panel of cross-sectional roadway data with a roadway section-specific serial correlation across time was explored. The negative multinomial model specification is compared with previous research, which used the same database but which also used negative binomial and random-effects negative binomial count models. If there is no section-specific correlation in the panel, the negative multinomial model becomes equivalent to the negative binomial. The differences in the estimation results between those models show that such a correlation exists in the data. The results show that the negative multinomial significantly outperforms the negative binomial and the random-effects negative binomial in terms of fit, with a statistically significantly higher likelihood at convergence. The signs of the coefficients were similar hi all models; when the signs differed, the negative multinomial model results were more intuitive. Overall, the analysis supports the use of the negative multinomial count model to estimate median crossover accident frequency models that are based on panel data.

AB - The use of the negative multinomial model to form a predictive model of median crossover accident frequencies with a multiyear panel of cross-sectional roadway data with a roadway section-specific serial correlation across time was explored. The negative multinomial model specification is compared with previous research, which used the same database but which also used negative binomial and random-effects negative binomial count models. If there is no section-specific correlation in the panel, the negative multinomial model becomes equivalent to the negative binomial. The differences in the estimation results between those models show that such a correlation exists in the data. The results show that the negative multinomial significantly outperforms the negative binomial and the random-effects negative binomial in terms of fit, with a statistically significantly higher likelihood at convergence. The signs of the coefficients were similar hi all models; when the signs differed, the negative multinomial model results were more intuitive. Overall, the analysis supports the use of the negative multinomial count model to estimate median crossover accident frequency models that are based on panel data.

UR - http://www.scopus.com/inward/record.url?scp=1842532538&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1842532538&partnerID=8YFLogxK

U2 - 10.3141/1840-22

DO - 10.3141/1840-22

M3 - Article

AN - SCOPUS:1842532538

SP - 193

EP - 197

JO - Transportation Research Record

JF - Transportation Research Record

SN - 0361-1981

IS - 1840

ER -