Estimating Dispersion Parameter of Negative Binomial Distribution for Analysis of Crash Data

The objective of this study is to improve the estimation of the dispersion parameter of the negative binomial distribution for modeling motor vehicle collisions. The negative binomial distribution is widely used to model count data such as traffic crash data, which often exhibit low sample mean values and small sample sizes. Under such situations, the most commonly used methods for estimating the dispersion parameter–the method of moment and the maximum likelihood estimate–may become inaccurate and unstable. A bootstrapped maximum likelihood method is proposed to improve the estimation of the dispersion parameter. The proposed method combines the technique of bootstrap resampling with the maximum likelihood estimation method to obtain better estimates of the dispersion parameter. The performance of the bootstrapped maximum likelihood estimate is compared with the method of moment and the maximum likelihood estimates through Monte Carlo simulations. To validate the simulation results, the methods are applied to observed data collected at four-leg unsignalized intersections in Toronto, Ontario, Canada. Overall, the results show that the proposed bootstrap maximum likelihood method produces smaller biases and more stable estimates. The improvements are more pronounced with small samples and low sample means.

[1]  G R Wood.  Confidence and prediction intervals for generalised linear accident models. , 2005, Accident; analysis and prevention.

[2]  H. Mantel,et al.  VARIANCE ESTIMATION FOR THE NATIONAL POPULATION HEALTH SURVEY , 2002 .

[3]  M J Maher,et al.  A comprehensive methodology for the fitting of predictive accident models. , 1996, Accident; analysis and prevention.

[4]  L. R. Taylor,et al.  Aggregation, Variance and the Mean , 1961, Nature.

[5]  R. Rosychuk,et al.  Comparison of variance estimation approaches in a two‐state Markov model for longitudinal data with misclassification , 2006, Statistics in medicine.

[6]  Fred L. Mannering,et al.  Negative binomial analysis of intersection accident frequencies , 1996 .

[7]  Nicholas I. Fisher,et al.  Bootstrap Confidence Regions for Directional Data , 1989 .

[8]  Abdelhak M. Zoubir.  Multiple bootstrap tests and their application , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  J. L. Folks,et al.  Multistage estimation compared with fixed-sample-size estimation of the negative binomial parameter k , 1984 .

[10]  B. Efron,et al.  The Jackknife: The Bootstrap and Other Resampling Plans. , 1983 .

[11]  Jerzy Neyman,et al.  On a New Class of "Contagious" Distributions, Applicable in Entomology and Bacteriology , 1939 .

[12]  Henrik Wann Jensen,et al.  Adaptive Smpling and Bias Estimation in Path Tracing , 1997, Rendering Techniques.

[13]  F. J. Anscombe,et al.  Sampling theory of the negative binomial and logarithmic series distributions. , 1950, Biometrika.

[14]  Simon Washington,et al.  Validation of FHWA Crash Models for Rural Intersections: Lessons Learned , 2003 .

[15]  Susan R. Wilson,et al.  Two guidelines for bootstrap hypothesis testing , 1991 .

[16]  C. I. Bliss,et al.  NEGATIVE BINOMIAL DISTRIBUTIONS WITH A COMMON K , 1958 .

[17]  Léopold Simar,et al.  Computer Intensive Methods in Statistics , 1994 .

[18]  Anscombe Fj.  The statistical analysis of insect counts based on the negative binomial distribution. , 1949 .

[19]  Dominique Lord,et al.  The prediction of accidents on digital networks, characteristics and issues related to the application of accident prediction models , 2000 .

[20]  R C Peck,et al.  Are accidents Poisson distributed? A statistical test. , 1994, Accident; analysis and prevention.

[21]  Timothy C. Coburn,et al.  Statistical and Econometric Methods for Transportation Data Analysis , 2004, Technometrics.

[22]  Joe N. Perry,et al.  Estimation of the Negative Binomial Parameter κ by Maximum Quasi -Likelihood , 1989 .

[23]  Abdel H. El-Shaarawi,et al.  Negative Binomial Distribution , 2006 .

[24]  W. Piegorsch.  Maximum likelihood estimation for the negative binomial dispersion parameter. , 1990, Biometrics.

[25]  Boualem Boashash,et al.  The bootstrap applied to passive acoustic aircraft parameter estimation , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[26]  Nils Toft,et al.  The Gamma-Poisson model as a statistical method to determine if micro-organisms are randomly distributed in a food matrix. , 2006, Food microbiology.

[27]  E Hauer,et al.  Overdispersion in modelling accidents on road sections and in empirical bayes estimation. , 2001, Accident; analysis and prevention.

[28]  Luis F. Miranda-Moreno,et al.  Effects of Low Sample Mean Values and Small Sample Size on Estimation of Fixed Dispersion Parameter of Poisson-Gamma Models for Motor Vehicle Crashes: Bayesian Perspective , 2007 .

[29]  Angelo J. Canty.  Bootstrap Techniques for Signal Processing , 2007 .

[30]  Ezra Hauer,et al.  OBSERVATIONAL BEFORE-AFTER STUDIES IN ROAD SAFETY -- ESTIMATING THE EFFECT OF HIGHWAY AND TRAFFIC ENGINEERING MEASURES ON ROAD SAFETY , 1997 .

[31]  Dominique Lord,et al.  Modeling motor vehicle crashes using Poisson-gamma models: examining the effects of low sample mean values and small sample size on the estimation of the fixed dispersion parameter. , 2006, Accident; analysis and prevention.

[32]  J. Lawless.  Negative binomial and mixed Poisson regression , 1987 .

[33]  J. H. Matis,et al.  Small Sample Comparison of Different Estimators of Negative Binomial Parameters , 1977 .

[34]  G Maycock,et al.  ACCIDENTS AT 4-ARM ROUNDABOUTS , 1984 .

[35]  C. I. Bliss,et al.  FITTING THE NEGATIVE BINOMIAL DISTRIBUTION TO BIOLOGICAL DATA AND NOTE ON THE EFFICIENT FITTING OF THE NEGATIVE BINOMIAL , 1953 .

[36]  Bhagwant Persaud,et al.  Empirical Bayes before-after safety studies: lessons learned from two decades of experience and future directions. , 2007, Accident; analysis and prevention.

[37]  B. Efron.  Nonparametric estimates of standard error: The jackknife, the bootstrap and other methods , 1981 .

[38]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

[39]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .

[40]  Hoong Chor Chin,et al.  Identification of Accident Causal Factors and Prediction of Hazardousness of Intersection Approaches , 2003 .

[41]  R Kulmala,et al.  Measuring the contribution of randomness, exposure, weather, and daylight to the variation in road accident counts. , 1995, Accident; analysis and prevention.

[42]  B. Efron.  The jackknife, the bootstrap, and other resampling plans , 1987 .

[43]  Bradley Efron.  2. The Jackknife Estimate of Bias , 1982 .

[44]  James A Bonneson,et al.  Calibration of Predictive Models for Estimating Safety of Ramp Design Configurations , 2005 .

[45]  Dimitris N. Politis,et al.  Computer-intensive methods in statistical analysis , 1998, IEEE Signal Process. Mag..