Computer based statistical treatment in models with incidental parameters: inspired by car crash data

Download
Ola Hössjer

Bootstrap and Markov chain Monte Carlo methods have received much attention in recent years. We study computer intensive methods that can be used in complex situations where it is not possible to express the likelihood estimates or the posterior analytically. The work is inspired by a set of car crash data from real traffic.

We formulate and develop a model for car crash data that aims to estimate and compare the relative collision safety among different car models. This model works sufficiently well, although complications arise due to a growing vector of incidental parameters. The bootstrap is shown to be a useful tool for studying uncertainties of the estimates of the structural parameters. This model is further extended to include driver characteristics. In a Poisson model with similar, but simpler structure, estimates of the structural parameter in the presence of incidental parameters are studied. The profile likelihood, bootstrap and the delta method are compared for deterministic and random incidental parameters. The same asymptotic properties, up to first order, are seen for deterministic as well as random incidental parameters.

The search for suitable methods that work in complex model structures leads us to consider Markov chain Monte Carlo (MCMC) methods. In the area of MCMC, we consider particularly the question of how and when to claim convergence of the MCMC run in situations where it is only possible to analyse the output values of the run and also how to compare different MCMC modellings. In Metropolis-Hastings algorithm, different proposal functions lead to different realisations. We develop a new convergence diagnostic, based on the Kullback-Leibler distance, which is shown to be particularly useful when comparing different runs. Comparisons with established methods turn out favourably for the KL.

In both models, a Bayesian analysis is made where the posterior distribution is obtained by MCMC methods. The credible intervals are compared to the corresponding confidence intervals from the bootstrap analysis and are shown to give the same qualitative conclusions.

MEET US


5-6
Dec

Vision Zero for Sustainable Road Safety in the Baltic Sea Region

The conference Vision Zero for Sustainable Road Safety in the Baltic Sea Region is held i Vilnius, Lithuania. VTI is part of the organisation committé and also one of the speakers.
25-27
Aug

ICTTP 2020

ICTTP, International Conference on Traffic and Transport Psychology, is held in Gothenburg, Sweden.

LATEST NEWS


2018-10-29

International standardisation efforts have many advantages

VTI participates in several international standardisation committees. The work is important because it helps to ensure that standards can be adapted to Swedish conditions and it also provides access to valuable contacts and networks.


2018-10-24

China wants to work with the best

Through the CTS cooperation, VTI is gaining valuable research contacts with China. The country is facing major challenges in the field of road safety but also has enormous potential.


2018-10-23

VTI participated in conference on electric roads

Systems with electrified roads are a relatively new concept and many projects have been launched in recent years. To stimulate the transfer of knowledge and collaboration, the Research and Innovation Platform for Electric Roads arranged its second...


2018-10-18

ADAS&ME is tackling the interaction between people and technology

ADAS&ME is a major EU project focused on automation, the human condition and the human environment. The budget is EUR 9.6 million and VTI is the coordinator.


2018-10-05

Users contribute to the development of train simulators

Apart from advanced driving simulators, VTI has developed several variations of train simulators which are used for training, education and research. In recent years, interest has increased drastically among major actors in the railway sector, and VTI has...


2018-10-03

VR study to contribute to a better working environment for bus drivers

A study where bus drivers test autonomous driving in a VR environment may contribute to a better working environment with reduced stress on the driver and safer driving.