A Bayesian finite mixture change-point model for assessing the risk of novice teenage drivers. (English) Zbl 1516.62421
Summary: The driving risk during the initial period after licensure for novice teenage drivers is typically the highest but decreases rapidly right after. The change-point of driving risk is a critical parameter for evaluating teenage driving risk, which also varies substantially among drivers. This paper presents latent class recurrent-event change-point models for detecting the change-points. The proposed model is applied to the Naturalist Teenage Driving Study, which continuously recorded the driving data of 42 novice teenage drivers for 18 months using advanced in-vehicle instrumentation. We propose a hierarchical BFMM to estimate the change-points by clusters of drivers with similar risk profiles. The model is based on a non-homogeneous Poisson process with piecewise-constant intensity functions. Latent variables which identify the membership of the subjects are used to detect potential clusters among subjects. Application to the Naturalistic Teenage Driving Study identifies three distinct clusters with change-points at 52.30, 108.99 and 150.20 hours of driving after first licensure, respectively. The overall intensity rate and the pattern of change also differ substantially among clusters. The results of this research provide more insight in teenagers’ driving behaviour and will be critical to improve young drivers’ safety education and parent management programs, as well as provide crucial reference for the GDL regulations to encourage safer driving.
MSC:
| 62-XX | Statistics |
Keywords:
change-point models; latent class models; naturalistic driving study; recurrent events; teenage driving riskReferences:
| [1] | T. Ando, Bayesian Model Selection and Statistical Modeling (Statistics: A Series of Textbooks and Monographs), Chapman & Hall/CRC, Boca Raton, FL, 2010. · Zbl 1303.62006 · doi:10.1201/EBK1439836149 |
| [2] | C.E. Antoniak, Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems, Ann. Statist. 2 (1974), pp. 1152-1174. · Zbl 0335.60034 · doi:10.1214/aos/1176342871 |
| [3] | J.A. Aschar, S. Loibel, and M. Andrade, Interfailure data with constant hazard function in the presence of change-points, REVSTAT 5 (2007), pp. 209-226. · Zbl 1513.62207 |
| [4] | A. Berg, R. Meyer, and J. Yu, Deviance information criterion for comparing stochastic volatility models, J. Bus. Econom. Statist. 22 (2004), pp. 107-120. |
| [5] | J. Besag, P. Green, D. Higdon, and K. Mengersen, Bayesian computation and stochastic systems, Statist. Sci. 10 (1995), pp. 3-66. · Zbl 0955.62552 · doi:10.1214/ss/1177010123 |
| [6] | O. Cappé, C. Robert, and T. Rydén, Reversible jump MCMC converging to birth-and-death MCMC and more general continuous time samplers, J. Royal Statist. Soc. Ser. B 65 (2002), pp. 679-700. · Zbl 1063.62133 · doi:10.1111/1467-9868.00409 |
| [7] | G. Celeux, M. Hurn, and C.P. Robert, Computational and inferential difficulties with mixture posterior distributions, J. Amer. Statist. Assoc. 95 (2000), pp. 957-970. · Zbl 0999.62020 |
| [8] | Centers for Disease Control and Prevention. Leading causes of death reports 1999-2005, 2008, Available at http://webappa.cdc.gov/sasweb/ncipc/leadcaus10.html. |
| [9] | S. Clarke and T.I. Robertson, A meta-analytic review of the big five personality factors and accident involvement in occupational and non-occupational settings, J. Occup. Organ. Psychol. 78 (2005), pp. 355-376. · doi:10.1348/096317905X26183 |
| [10] | R.J. Cook and J.F. Lawless, The Statistical Analysis of Recurrent Events, Statistics for Biology and Health, Springer, New York, NY, 2007. · Zbl 1159.62061 |
| [11] | H.A. Deery and B.N. Fildes, Young novice driver subtypes: Relationship to high-risk behavior, traffic accident record, and simulator driving performance, Hum. Fact.: J. Hum. Fact. Ergon. Soc. 41 (1999), pp. 628-643. · doi:10.1518/001872099779656671 |
| [12] | J. Diebolt and C.P. Robert, Estimation of finite mixture distributions through Bayesian sampling, J. R. Statist. Soc. Ser. B 56 (1994), pp. 363-375. · Zbl 0796.62028 |
| [13] | T.A. Dingus, S.G. Klauer, V.L. Neale, A. Petersen, S.E. Lee, J. Sudweeks, M.A. Perez, J. Hankey, D. Ramsey, S. Gupta, C. Bucher, Z.R. Doerzaph, J. Jermeland, and R.R. Knipling, The 100-car naturalistic driving study: Phase ii – results of the 100-car field experiment, Tech. Rep. DOT-HS-810-593, National Highway Traffic Safety Administration, Washington, DC, 2006. |
| [14] | D. Frobish and N. Ebrahimi, Parametric estimation of change-points for actual event data in recurrent events models, Comput. Statist. Data Anal. 53 (2009), pp. 671-682. · Zbl 1452.62807 · doi:10.1016/j.csda.2008.08.022 |
| [15] | S. Frühwirth-Schnatter, Finite mixture and Markov switching models, Springer Series in Statistics, Springer, New York, NY, 2006. · Zbl 1108.62002 |
| [16] | A. Gelman, J.B. Carlin, H.S. Stern, and D.B. Rubin, Bayesian Data Analysis, 2nd ed., Texts in Statistical Science Series, Chapman & Hall/CRC, Boca Raton, FL, 2004. · Zbl 1039.62018 |
| [17] | A. Gelman and D.B. Rubin, Inference from iterative simulation using multiple sequences, Stat. Sci. 7 (1992), pp. 457-511. · Zbl 1386.65060 · doi:10.1214/ss/1177011136 |
| [18] | L.A. Goodman, Exploratory latent structure analysis using both identifiable and unidentifiable models, Biometrika 61 (1974), pp. 215-231. · Zbl 0281.62057 · doi:10.1093/biomet/61.2.215 |
| [19] | Governor’s Highway Safety Association. State laws & funding: Virginia, 2015, Available at http://www.ghsa.org/html/stateinfo/bystate/va.html. |
| [20] | P. Green, Reversible jump MCMC computation and Bayesian model determination, Biometrika 82 (1995), pp. 711-732. · Zbl 0861.62023 · doi:10.1093/biomet/82.4.711 |
| [21] | M. Gruet, A. Philippe, and C. Robert, MCMC control spreadsheets for exponential mixture estimation, J. Comput. Graph. Stat. 8 (1999), pp. 298-317. |
| [22] | F. Guo and Y.J. Fang, Individual driver risk assessment using naturalistic driving data, Accid. Anal. Prev. 61 (2013), pp. 3-9. · doi:10.1016/j.aap.2012.06.014 |
| [23] | F. Guo, Y.J. Fang, and J.F. Antin, Evaluation of older driver fitness-to-drive metrics and driving risk using naturalistic driving study data, Tech. Rep. 15-UM-036, Federal Highway Administration, 2015. |
| [24] | F. Guo, Y.J. Fang, and J.F. Antin, Older driver fitness-to-drive evaluation using naturalistic driving data, J. Safety Res. 54 (2015), pp. 49-54. · doi:10.1016/j.jsr.2015.06.013 |
| [25] | F. Guo, S.G. Klauer, J.M. Hankey, and T.A. Dingus, Using near-crashes as a crash surrogate for naturalistic driving studies, J. Transport. Res. Board 2147 (2010), pp. 66-74. · doi:10.3141/2147-09 |
| [26] | F. Guo, B.G. Simons-Morton, S.E. Klauer, M.C. Ouimet, T.A. Dingus, and S.E. Lee, Variability in crash and near-crash risk among novice teenage drivers: a naturalistic study, J. Pediatrics 163 (2013), pp. 1670-1676. · doi:10.1016/j.jpeds.2013.07.025 |
| [27] | J.A. Hagenaars and A.L. McCutcheon (eds.), Applied Latent Class Analysis, Cambridge University Press, Cambridge, 2002. · Zbl 1003.00021 · doi:10.1017/CBO9780511499531 |
| [28] | H. Ishwaran, L.F. James, and J.Y. Sun, Bayesian model selection in finite mixtures by marginal density decompositions, J. Amer. Statist. Assoc. 96 (2001), pp. 1316-1332. · Zbl 1051.62027 |
| [29] | D.S. Karasoy and C. Kadilar, A new bayes estimate of change-point in the hazard function, Comput. Statist. Data Anal. 51 (2007), pp. 2993-3001. · Zbl 1161.62341 · doi:10.1016/j.csda.2006.07.007 |
| [30] | R.E. Kass and A.E. Raftery, Bayes factors and model uncertainty, Tech. Rep., University of Washington, Department of Statistics, Seattle, WA, 1993. |
| [31] | S. Kim, Z. Chen, Z. Zhang, B.G. Simons-Morton, and P.S. Albert, Bayesian hierarchical Poisson regression models: an application to a driving study with kinematic events, J. Amer. Statist. Assoc. 108 (2013), pp. 494-503. · Zbl 06195955 |
| [32] | S.G. Klauer, T.A. Dingus, V.L. Neale, J. Sudweeks, and D.J. Ramsey, Comparing real-world behaviors of drivers with high vs. low rates of crashes and near-crashes, Tech. Rep. DOT-HS-811-091, National Highway Traffic Safety Administration, Washington, DC, 2009. |
| [33] | S.G. Klauer, F. Guo, B.G. Simons-Morton, M.C. Ouimet, S.E. Lee, and T.A. Dingus, Distracted driving and risk of road crashes among novice and experienced drivers, New Engl. J. Med. 370 (2014), pp. 54-59. · doi:10.1056/NEJMsa1204142 |
| [34] | S.G. Klauer, F. Guo, J.D. Sudweeks, and T.A. Dingus, An analysis of driver inattention using a case-crossover approach on 100-car data, Tech. Rep. DOT-HS-811-334, National Highway Traffic Safety Administration, Washington, DC, 2010. |
| [35] | R.W. Klein and S.D. Roberts, A time-varying Poisson arrival process generator, Simulation 43 (1984), pp. 193-195. · doi:10.1177/003754978404300406 |
| [36] | S. Kotz, N. Balakrishnan, and N.L. Johnson, Continuous Multivariate Distributions. Volume 1, John Wiley, New York, NY, 2000. · Zbl 0946.62001 · doi:10.1002/0471722065 |
| [37] | J. Lawless and M. Zhan, Analysis of interval-grouped recurrent-event data using piecewise constant rate functions, Canad. J. Stat. 26 (1998), pp. 549-565. · Zbl 0963.62063 · doi:10.2307/3315717 |
| [38] | P. Lazarsfeld and N. Henry, Latent Structure Analysis, John Wiley, New York, NY, 1968. · Zbl 0182.52201 |
| [39] | S.E. Lee, B.G. Simons-Morton, S.E. Klauer, M.C. Ouimet, and T.A. Dingus, Naturalistic assessment of novice teenage crash experience, Accid. Anal. Prev. 43 (2011), pp. 1472-1479. · doi:10.1016/j.aap.2011.02.026 |
| [40] | Q. Li, F. Guo, S. Klauer, and B. Simons-Monton, Recurrent-event change-point models for novice teenage driving risk evaluation, Comput. Statist. Data Anal., submitted, 2016. |
| [41] | J.S. Liu, The collapsed Gibbs sampler in Bayesian computations with applications to a gene regulation problem, J. Amer. Statist. Assoc. 89 (2003), pp. 958-966. · Zbl 0804.62033 |
| [42] | C.R. Loader, Inference for a hazard rate change-point, Biometrika 78 (1991), pp. 749-757. · Zbl 0752.62079 · doi:10.1093/biomet/78.4.749 |
| [43] | J.M. Marin, K. Mengersen, and C.P. Robert, Bayesian modelling and inference on mixtures of distributions, in Bayesian Thinking: Modeling and computation, Handbook of Statistics, Vol. 25, C. Rao and D. Dey, eds., Elsevier/North-Holland, Amsterdam, 2005, pp. 459-507. · Zbl 1136.62012 |
| [44] | D. Matthews and V. Farewell, On testing for a constant hazard against a change-point alternative, Biometrics 38 (1982), pp. 463-468. · doi:10.2307/2530460 |
| [45] | D.R. Mayhew, H.M. Simpson, and A. Pak, Changes in collision rates among novice drivers during the first months of driving, Accid. Anal. Prev. 35 (2003), pp. 683-691. · doi:10.1016/S0001-4575(02)00047-7 |
| [46] | A.C. McCutcheon, Latent Class Analysis, Sage Publications, Newbury Park, CA, 1987. · doi:10.4135/9781412984713 |
| [47] | G. McLachlan and D. Peel, Finite Mixture Models, Wiley Series in Probability and Statistics: Applied Probability and Statistics, Wiley-Interscience, New York, 2000. · Zbl 0963.62061 · doi:10.1002/0471721182 |
| [48] | K. Mengersen and C. Robert, Testing for mixtures: A Bayesian entropic approach (with discussion), Bayesian Statistics, Vol. 5, Oxford University Press, Oxford, 1996. |
| [49] | H.G. Müller and J.L. Wang, Change-point models for hazard functions, in Change-Point Problems, IMS Lecture Notes Monogr. Ser., Vol. 23, E. Carlstein, H.G. Muller, and D. Siegmund, eds., Institute of Mathematical Statistics, Hayward, CA, 1994, pp. 224-241. · Zbl 1163.62349 |
| [50] | K.P. Murphy, Machine Learning: A Probabilistic Perspective, MIT Press, Boston, MA, 2012. · Zbl 1295.68003 |
| [51] | C. Musselwhite, Attitudes towards vehicle driving behavior: Categorizing and contextualizing risk, Accid. Anal. Prev. 38 (2006), pp. 324-334. · doi:10.1016/j.aap.2005.10.003 |
| [52] | National Highway Traffic Safety Administration: Traffic safety facts 2000: Young drivers, Tech. Rep. DOT-HS-809-336, National Center for Statistics & Analysis, Washington, DC, 2000. |
| [53] | R.M. Neal, Slice sampling, Ann. Statist. 31 (2003), pp. 705-767. · Zbl 1051.65007 · doi:10.1214/aos/1056562461 |
| [54] | A. Nobile, Bayesian finite mixtures: A note on prior specification and posterior computation, Tech. Rep. 05-3, University of Glasgow, Department of Statistics, Scotland, UK, 2007. |
| [55] | M.C. Ouimet, T.G. Brown, F. Guo, S.G. Klauer, B.G. Simons-Morton, Y.J. Fang, S.E. Lee, C. Gianoulakis, and T.A. Dingus, Higher crash and near-crash rates in teenaged drivers with lower cortisol response: An 18-month longitudinal, naturalistic study, JAMA Pediatrics 168 (2014), pp. 517-522. · doi:10.1001/jamapediatrics.2013.5387 |
| [56] | S. Richardson and P.J. Green, On Bayesian analysis of mixtures with an unknown number of components (with discussion), J. Roy. Statist. Soc. Ser. B 59 (1997), pp. 731-792. · Zbl 0891.62020 · doi:10.1111/1467-9868.00095 |
| [57] | C.P. Robert, The Bayesian Choice, 2nd ed., Springer, New York, NY, 2001. · Zbl 0980.62005 |
| [58] | K. Roeder and L. Wasserman, Practical Bayesian density estimation using mixtures of normals, J. Amer. Statist. Assoc. 92 (1997), pp. 894-902. · Zbl 0889.62021 |
| [59] | G. Rolls, R.D. Hall, R. Ingham, and M. McDonald, Accident Risk and Behavioral Patterns of Younger Drivers, AA Foundation for Road Safety Research, Hampshire, UK, 1991. |
| [60] | S.M. Ross, Simulation, Academic Press, San Diego, CA, 2006. · Zbl 1111.65008 |
| [61] | S.K. Sahu and R.C.H. Cheng, A fast distance-based approach for determining the number of components in mixtures, Canad. J. Statist. 31 (2003), pp. 3-22. · Zbl 1039.62022 · doi:10.2307/3315900 |
| [62] | G. Schwarz, Estimating the dimension of a model, Ann. Statist. 6 (1978), pp. 461-464. · Zbl 0379.62005 · doi:10.1214/aos/1176344136 |
| [63] | B.G. Simons-Morton, M.C. Ouimet, Z. Zhang, S.E. Lee, S.E. Klauer, J. Wang, P.S. Albert, and T.A. Dingus, Crash and risky driving involvement among novice adolescent drivers and their parents, Amer. J. Pub. Health 101 (2011), pp. 2362-2367. · doi:10.2105/AJPH.2011.300248 |
| [64] | B.G. Simons-Morton, M.C. Ouimet, Z. Zhang, S.E. Lee, S.E. Klauer, J. Wang, R. Chen, P.S. Albert, and T.A. Dingus, The effect of passengers and risk-taking friends on risky driving and crashes/near crashes among novice teenagers, J. Adolesc. Health. 49 (2011), pp. 587-593. · doi:10.1016/j.jadohealth.2011.02.009 |
| [65] | D.J. Spiegelhalter, N.G. Best, B.P. Carlin, and A. Van Der Linde, Bayesian measures of model complexity and fit, J. Roy. Statist. Soc. Ser. B 64 (2002), pp. 583-639. · Zbl 1067.62010 · doi:10.1111/1467-9868.00353 |
| [66] | M. Stephens, Bayesian analysis of mixture models with an unknown number of components – an alternative to reversible jump methods, Ann. Statist. 28 (2000), pp. 40-74. · Zbl 1106.62316 · doi:10.1214/aos/1016120364 |
| [67] | M. Stephens, Dealing with label switching in mixture models, J. R. Stat. Soc. Ser. B Stat. Methodol. 62 (2000), pp. 795-809. · Zbl 0957.62020 · doi:10.1111/1467-9868.00265 |
| [68] | Y.W. Teh, Encyclopedia of Machine Learning, Springer, New York. |
| [69] | W. Thompson, Point Process Models with Applications to Safety and Reliability, Chapman and Hall, New York, NY, 1988. · Zbl 0733.60069 · doi:10.1007/978-1-4613-1067-9 |
| [70] | D.M. Titterington, A.F.M. Smith, and U.E. Makov, Statistical Analysis of Finite Mixture Distributions, Wiley Series in Probability and Mathematical Statistics: Applied Probability and Statistics, John Wiley, Chichester, 1985. · Zbl 0646.62013 |
| [71] | P. Ulleberg, Personality subtypes of young drivers. relationship to risk-taking preferences, accident involvement, and response to a traffic safety campaign, Transp. Res. F: Traffic Psychol. Behav. 4 (2001), pp. 279-297. · doi:10.1016/S1369-8478(01)00029-8 |
| [72] | R. West and R. Odgen, Continuous-time estimation of a change-point in a Poisson process, JSCS 56 (1997), pp. 293-302. · Zbl 0872.62082 |
| [73] | A.F. Williams, Teenage drivers: Patterns of risk, J. Safety Res. 34 (2003), pp. 5-15. · doi:10.1016/S0022-4375(02)00075-0 |
| [74] | Y.C. Yao, Maximum likelihood estimation in hazard rate models with a change-point, Commun. Statist. A. Theory Methods 15 (1986), pp. 2455-2466. · Zbl 0606.62109 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
