An efficient parallel sampling technique for Multivariate Poisson-Lognormal model: Analysis with two crash count datasets

Zhan, Xianyuan; Aziz, H. M. Abdul; Ukkusuri, Satish V.

doi:10.1016/j.amar.2015.10.002

Title: An efficient parallel sampling technique for Multivariate Poisson-Lognormal model: Analysis with two crash count datasets

Journal Article · Thu Nov 19 00:00:00 EST 2015 · Analytic Methods in Accident Research

DOI:https://doi.org/10.1016/j.amar.2015.10.002· OSTI ID:1265895

Zhan, Xianyuan ^[1]; Aziz, H. M. Abdul ^[1]; Ukkusuri, Satish V. ^[1]

Purdue Univ., West Lafayette, IN (United States)

Our study investigates the Multivariate Poisson-lognormal (MVPLN) model that jointly models crash frequency and severity accounting for correlations. The ordinary univariate count models analyze crashes of different severity level separately ignoring the correlations among severity levels. The MVPLN model is capable to incorporate the general correlation structure and takes account of the over dispersion in the data that leads to a superior data fitting. But, the traditional estimation approach for MVPLN model is computationally expensive, which often limits the use of MVPLN model in practice. In this work, a parallel sampling scheme is introduced to improve the original Markov Chain Monte Carlo (MCMC) estimation approach of the MVPLN model, which significantly reduces the model estimation time. Two MVPLN models are developed using the pedestrian vehicle crash data collected in New York City from 2002 to 2006, and the highway-injury data from Washington State (5-year data from 1990 to 1994) The Deviance Information Criteria (DIC) is used to evaluate the model fitting. The estimation results show that the MVPLN models provide a superior fit over univariate Poisson-lognormal (PLN), univariate Poisson, and Negative Binomial models. Moreover, the correlations among the latent effects of different severity levels are found significant in both datasets that justifies the importance of jointly modeling crash frequency and severity accounting for correlations.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 1265895

Journal Information:: Analytic Methods in Accident Research, Vol. 8, Issue C; ISSN 2213-6657

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (35)

Full Bayes Poisson gamma, Poisson lognormal, and zero inflated random effects models: Comparing the precision of crash frequency estimates Aguero-Valverde, Jonathan Accident Analysis & Prevention, Vol. 50 https://doi.org/10.1016/j.aap.2012.04.019	journal	January 2013
Bayesian Multivariate Poisson Lognormal Models for Crash Severity Modeling and Site Ranking Aguero-Valverde, Jonathan; Jovanis, Paul P. Transportation Research Record: Journal of the Transportation Research Board, Vol. 2136, Issue 1 https://doi.org/10.3141/2136-10	journal	January 2009
Spatial Correlation in Multilevel Crash Frequency Models: Effects of Different Neighboring Structures Aguero-Valverde, Jonathan; Jovanis, Paul P. Transportation Research Record: Journal of the Transportation Research Board, Vol. 2165, Issue 1 https://doi.org/10.3141/2165-03	journal	January 2010
Exploring the determinants of pedestrian–vehicle crash severity in New York City Aziz, H. M. Abdul; Ukkusuri, Satish V.; Hasan, Samiul Accident Analysis & Prevention, Vol. 50 https://doi.org/10.1016/j.aap.2012.09.034	journal	January 2013
A Full Bayesian multivariate count data model of collision severity with spatial correlation Barua, Sudip; El-Basyouny, Karim; Islam, Md Tazul Analytic Methods in Accident Research, Vol. 3-4 https://doi.org/10.1016/j.amar.2014.09.001	journal	October 2014
The covariance between the number of accidents and the number of victims in multivariate analysis of accident related outcomes Bijleveld, F. D. Accident Analysis & Prevention, Vol. 37, Issue 4 https://doi.org/10.1016/j.aap.2005.01.004	journal	July 2005
A crash-prediction model for road tunnels Caliendo, Ciro; De Guglielmo, Maria Luisa; Guida, Maurizio Accident Analysis & Prevention, Vol. 55 https://doi.org/10.1016/j.aap.2013.02.024	journal	June 2013
A latent variable representation of count data models to accommodate spatial and temporal dependence: Application to predicting crash frequency at intersections Castro, Marisol; Paleti, Rajesh; Bhat, Chandra R. Transportation Research Part B: Methodological, Vol. 46, Issue 1 https://doi.org/10.1016/j.trb.2011.09.007	journal	January 2012
Markov Chain Monte Carlo Analysis of Correlated Count Data Chib, Siddhartha; Winkelmann, Rainer Journal of Business & Economic Statistics, Vol. 19, Issue 4 https://doi.org/10.1198/07350010152596673	journal	October 2001
Modeling crash frequency and severity using multinomial-generalized Poisson model with error components Chiou, Yu-Chiun; Fu, Chiang Accident Analysis & Prevention, Vol. 50 https://doi.org/10.1016/j.aap.2012.03.030	journal	January 2013
Traffic accident segmentation by means of latent class clustering Depaire, Benoît; Wets, Geert; Vanhoof, Koen Accident Analysis & Prevention, Vol. 40, Issue 4 https://doi.org/10.1016/j.aap.2008.01.007	journal	July 2008
Collision prediction models using multivariate Poisson-lognormal regression El-Basyouny, Karim; Sayed, Tarek Accident Analysis & Prevention, Vol. 41, Issue 4 https://doi.org/10.1016/j.aap.2009.04.005	journal	July 2009
A latent class modeling approach for identifying vehicle driver injury severity factors at highway-railway crossings Eluru, Naveen; Bagheri, Morteza; Miranda-Moreno, Luis F. Accident Analysis & Prevention, Vol. 47 https://doi.org/10.1016/j.aap.2012.01.027	journal	July 2012
Econometric Models for Count Data with an Application to the Patents-R & D Relationship Hausman, Jerry; Hall, Bronwyn H.; Griliches, Zvi Econometrica, Vol. 52, Issue 4 https://doi.org/10.2307/1911191	journal	July 1984
Motor Vehicle Crashes and Land Use: Empirical Analysis from Hawaii Kim, Karl; Yamashita, Eric Transportation Research Record: Journal of the Transportation Research Board, Vol. 1784, Issue 1 https://doi.org/10.3141/1784-10	journal	January 2002
Demographic and environmental correlates of pedestrian injury collisions: a spatial analysis LaScala, Elizabeth A.; Gerber, Daniel; Gruenewald, Paul J. Accident Analysis & Prevention, Vol. 32, Issue 5 https://doi.org/10.1016/S0001-4575(99)00100-1	journal	September 2000
Death on the Crosswalk: A Study of Pedestrian-Automobile Collisions in Los Angeles Loukaitou-Sideris, Anastasia; Liggett, Robin; Sung, Hyun-Gun Journal of Planning Education and Research, Vol. 26, Issue 3 https://doi.org/10.1177/0739456X06297008	journal	March 2007
A multivariate Poisson-lognormal regression model for prediction of crash counts by severity, using Bayesian methods Ma, Jianming; Kockelman, Kara M.; Damien, Paul Accident Analysis & Prevention, Vol. 40, Issue 3 https://doi.org/10.1016/j.aap.2007.11.002	journal	May 2008
Analytic methods in accident research: Methodological frontier and future directions Mannering, Fred L.; Bhat, Chandra R. Analytic Methods in Accident Research, Vol. 1 https://doi.org/10.1016/j.amar.2013.09.001	journal	January 2014
Highway accident severities and the mixed logit model: An exploratory empirical analysis Milton, John C.; Shankar, Venky N.; Mannering, Fred L. Accident Analysis & Prevention, Vol. 40, Issue 1 https://doi.org/10.1016/j.aap.2007.06.006	journal	January 2008
A clustering regression approach: A comprehensive injury severity analysis of pedestrian–vehicle crashes in New York, US and Montreal, Canada Mohamed, Mohamed Gomaa; Saunier, Nicolas; Miranda-Moreno, Luis F. Safety Science, Vol. 54 https://doi.org/10.1016/j.ssci.2012.11.001	journal	April 2013
On accommodating spatial dependence in bicycle and pedestrian injury counts by severity level Narayanamoorthy, Sriram; Paleti, Rajesh; Bhat, Chandra R. Transportation Research Part B: Methodological, Vol. 55 https://doi.org/10.1016/j.trb.2013.07.004	journal	September 2013
Regression in a copula model for bivariate count data Nikoloulopoulos, Aristidis K.; Karlis, Dimitris Journal of Applied Statistics, Vol. 37, Issue 9 https://doi.org/10.1080/02664760903093591	journal	September 2010
Multivariate Poisson-Lognormal Models for Jointly Modeling Crash Frequency by Severity Park, Eun Sug; Lord, Dominique Transportation Research Record: Journal of the Transportation Research Board, Vol. 2019, Issue 1 https://doi.org/10.3141/2019-01	journal	January 2007
A fully Bayesian multivariate approach to before–after safety evaluation Park, Eun Sug; Park, Jaebeom; Lomax, Timothy J. Accident Analysis & Prevention, Vol. 42, Issue 4 https://doi.org/10.1016/j.aap.2009.12.026	journal	July 2010
A joint-probability approach to crash prediction models Pei, Xin; Wong, S. C.; Sze, N. N. Accident Analysis & Prevention, Vol. 43, Issue 3 https://doi.org/10.1016/j.aap.2010.12.026	journal	May 2011
Bayesian multivariate spatial models for roadway traffic crash mapping Song, J. J.; Ghosh, M.; Miaou, S. Journal of Multivariate Analysis, Vol. 97, Issue 1 https://doi.org/10.1016/j.jmva.2005.03.007	journal	January 2006
Bayesian measures of model complexity and fit Spiegelhalter, David J.; Best, Nicola G.; Carlin, Bradley P. Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 64, Issue 4 https://doi.org/10.1111/1467-9868.00353	journal	October 2002
Random Parameter Model Used to Explain Effects of Built-Environment Characteristics on Pedestrian Crash Frequency Ukkusuri, Satish; Hasan, Samiul; Aziz, H. M. Abdul Transportation Research Record: Journal of the Transportation Research Board, Vol. 2237, Issue 1 https://doi.org/10.3141/2237-11	journal	January 2011
Predicting accident frequency at their severity levels and its application in site ranking using a two-stage mixed multivariate model Wang, Chao; Quddus, Mohammed A.; Ison, Stephen G. Accident Analysis & Prevention, Vol. 43, Issue 6 https://doi.org/10.1016/j.aap.2011.05.016	journal	November 2011
A Poisson-lognormal conditional-autoregressive model for multivariate spatial analysis of pedestrian crash counts across neighborhoods Wang, Yiyi; Kockelman, Kara M. Accident Analysis & Prevention, Vol. 60 https://doi.org/10.1016/j.aap.2013.07.030	journal	November 2013
Seemingly Unrelated Negative Binomial Regression Winkelmann, Rainer Oxford Bulletin of Economics and Statistics, Vol. 62, Issue 4 https://doi.org/10.1111/1468-0084.00188	journal	September 2000
The heterogeneous effects of guardian supervision on adolescent driver-injury severities: A finite-mixture random-parameters approach Xiong, Yingge; Mannering, Fred L. Transportation Research Part B: Methodological, Vol. 49 https://doi.org/10.1016/j.trb.2013.01.002	journal	March 2013
Alternative Ordered Response Frameworks for Examining Pedestrian Injury Severity in New York City Yasmin, Shamsunnahar; Eluru, Naveen; Ukkusuri, Satish V. Journal of Transportation Safety & Security, Vol. 6, Issue 4 https://doi.org/10.1080/19439962.2013.839590	journal	May 2014
A simultaneous equations model of crash frequency by collision type for rural intersections Ye, Xin; Pendyala, Ram M.; Washington, Simon P. Safety Science, Vol. 47, Issue 3 https://doi.org/10.1016/j.ssci.2008.06.007	journal	March 2009

Cited By (1)

A joint econometric framework for modeling crash counts by severity Yasmin, Shamsunnahar; Eluru, Naveen Transportmetrica A: Transport Science, Vol. 14, Issue 3 https://doi.org/10.1080/23249935.2017.1369469	journal	July 2017

Similar Records

Regression methods for binomial and Poisson distributed data

Conference · Sun Jan 01 00:00:00 EST 1984 · OSTI ID:1265895

Frome, E L

A bivariate mixed lognormal distribution with an analysis of rainfall data

Journal Article · Mon Feb 01 00:00:00 EST 1993 · Journal of Applied Meteorology; (United States) · OSTI ID:1265895

Shimizu, Kunio

The relationship between truck accidents and geometric design of road sections: Poisson versus negative binomial regressions

Conference · Thu Jul 01 00:00:00 EDT 1993 · OSTI ID:1265895

Miaou, Shaw-Pin

Related Subjects

97 MATHEMATICS AND COMPUTING
accident analysis
pedestrian crashes
severity models

Title: An efficient parallel sampling technique for Multivariate Poisson-Lognormal model: Analysis with two crash count datasets

Citation Formats

References (35)

Cited By (1)

Similar Records

Related Subjects