Jacques J.F. Commandeur

On statistical inference in time series analysis of the evolution of road safety

Data collected for building a road safety observatory usually include observations made sequentially through time. Examples of such data, called time series data, include annual (or monthly) number of road traffic accidents, traffic fatalities or vehicle kilometers driven in a country, as well as the corresponding values of safety performance indicators (e.g., data on speeding, seat belt use, alcohol use, etc.). Some commonly used statistical techniques imply assumptions that are often violated by the special properties of time series data, namely serial dependency among disturbances associated with the observations. The first objective of this paper is to demonstrate the impact of such violations to the applicability of standard methods of statistical inference, which leads to an under or overestimation of the standard error and consequently may produce erroneous inferences. Moreover, having established the adverse consequences of ignoring serial dependency issues, the paper aims to describe rigorous statistical techniques used to overcome them. In particular, appropriate time series analysis techniques of varying complexity are employed to describe the development over time, relating the accident-occurrences to explanatory factors such as exposure measures or safety performance indicators, and forecasting the development into the near future. Traditional regression models (whether they are linear, generalized linear or nonlinear) are shown not to naturally capture the inherent dependencies in time series data. Dedicated time series analysis techniques, such as the ARMA-type and DRAG approaches are discussed next, followed by structural time series models, which are a subclass of state space methods. The paper concludes with general recommendations and practice guidelines for the use of time series models in road safety research.

A general solution of the weighted orthonormal procrustes problem

A general solution for weighted orthonormal Procrustes problem is offered in terms of the least squares criterion. For the two-demensional case. this solution always gives the global minimum; for the general case, an algorithm is proposed that must converge, although not necessarily to the global minimum. In general, the algorithm yields a solution for the problem of how to fit one matrix to another under the condition that the dimensions of the latter matrix first are allowed to be transformed orthonormally and then weighted differentially, which is the task encountered in fitting analogues of the IDIOSCAL and INDSCAL models to a set of configurations.

Latent risk and trend models for the evolution of annual fatality numbers in 30 European countries.

In this paper a unified methodology is presented for the modelling of the evolution of road safety in 30 European countries. For each country, annual data of the best available exposure indicator and of the number of fatalities were simultaneously analysed with the bivariate latent risk time series model. This model is based on the assumption that the amount of exposure and the number of fatalities are intrinsically related. It captures the dynamic evolution in the fatalities as the product of the dynamic evolution in two latent trends: the trend in the fatality risk and the trend in the exposure to that risk. Before applying the latent risk model to the different countries it was first investigated and tested whether the exposure indicator at hand and the fatalities in each country were in fact related at all. If they were, the latent risk model was applied to that country; if not, a univariate local linear trend model was applied to the fatalities series only, unless the latent risk time series model was found to yield better forecasts than the univariate local linear trend model. In either case, the temporal structure of the unobserved components of the optimal model was established, and structural breaks in the trends related to external events were identified and captured by adding intervention variables to the appropriate components of the model. As a final step, for each country the optimally modelled developments were projected into the future, thus yielding forecasts for the number of fatalities up to and including 2020.

Five road safety education programmes for young adolescent pedestrians and cyclists: a multi-programme evaluation in a field setting

A practical approach was developed to assess and compare the effects of five short road safety education (RSE) programmes for young adolescents that does not rely on injury or crash data but uses self reported behaviour. Questionnaires were administered just before and about one month after participation in the RSE programmes, both to youngsters who had participated in a RSE programme, the intervention group, and to a comparable reference group of youngsters who had not, the reference group. For each RSE programme, the answers to the questionnaires in the pre- and post-test were checked for internal consistency and then condensed into a single safety score using categorical principal components analysis. Next, an analysis of covariance was performed on the obtained safety scores in order to compare the post-test scores of the intervention and reference groups, corrected for their corresponding pre-test scores. It was found that three out of five RSE programmes resulted in significantly improved self-reported safety behaviour. However, the proportions of participants that changed their behaviour relative to the reference group were small, ranging from 3% to 20%. Comparisons among programme types showed cognitive approaches not to differ in effect from programmes that used fear-appeal approaches. The method used provides a useful tool to assess and compare the effects of different education programmes on self-reported behaviour.

Intervention time series analysis of crime rates:The case of sentence reform in Virginia

Abstract We review the basic concepts of intervention analysis in the context of structural time series models and we apply this methodology to investigate the possible reduction in monthly crime rates reported from January 1984 up to and including December 2010 after Virginia abolished parole and reformed sentencing in January 1995. We find that the change in legislation has significantly reduced the burglary rates and to a lesser extent the murder rates in Virginia. The robustness of our results is investigated with an automatic detection of breaks procedure as well as with analyses of quarterly rather than monthly data.

Consensus molecular alignment based on generalized Procrustes analysis

One of the most serious problems in three-dimensional quantitative structure-activity relationship (3D-QSAR) studies is selection of an alignment rule for molecular super position of the compounds in the data set. In 3D-QSAR analyses of structure-activity data, a reference compound in a defined conformation is chosen, and all structures in the data set are aligned with the reference in a pairwise manner. In subsequent steps, conformation/alignment-dependent descriptors are computed for the compounds and compared to those of the reference. This approach gives much weight to the arbitrarily chosen reference molecule and can introduce a bias in the results. Here an alternative, and more general, approach to molecular alignment is presented that is based on Generalized Procrustes Analysis (GPA). The result is a consensus alignment that uses all molecules in the data set and avoids the bias introduced in the pairwise alignment strategy.

Analysing the development of road safety using demographic data

The purpose of this paper is to show that time series analyses of road safety and risk can be improved by using demographic data. We demonstrate that the distance travelled by drivers or riders of a certain age reflects the fluctuations over the years of the number of people of that age within the population. We further demonstrate that the change over time of per capita distance travelled, i.e. distance travelled per person, is often less subject to stochastic fluctuations, and therefore more smooth than the total distance travelled for drivers of that age. This smoothness is used to obtain forecasts of distance travelled, or to average out year-to-year fluctuations of data of distance travelled. Analysis of such data stratified by age group, gender or both reveals that, for most travel modes, per capita distance travelled is to a large extent constant or slowly changing over time. The consequences for the evaluation of risk, i.e. casualties per distance travelled, with and without the use of population data, are explored. Dutch data are used to illustrate the model concept. It is shown that the analyses and forecasts of distance travelled could gain substantially by incorporating demographic data, as compared to an analysis with data of distance travelled alone. The paper further shows that, for an analysis of risk and therefore for traffic safety forecasts in the absence of any data of distance travelled, stratified analysis of mortality, i.e. casualties per inhabitant, may be a reasonable alternative.

Structural Intervention Time Series Analysis of Crime Rates: The Impact of Sentence Reform in Virginia

We adopt a structural time series analysis to investigate the impact of parole abolition and sentence reform in Virginia on reported crime rates. The Commonwealth of Virginia abolished parole and reformed sentencing for all felony offences committed on or after January 1, 1995. To examine the impact of Virginias change in legislation on reported crime rates from 1995 onwards, we perform an intervention time series analysis based on structural time series models. We empirically find that the change in legislation has significantly reduced the burglary rates and to a lesser extent the murder rates in Virginia. For other violent crimes such as rape and aggravated assault the evidence of a significant reduction in crime rates is less evident or is not found. This empirical study for Virginia also provides an illustration of how an effective intervention time series analysis can be carried out in crime studies.

Study on serious road traffic injuries in the EU

The general objective of this study is to collect knowledge that will enable the future identification of measures for effective prevention of serious road traffic injuries. The specific objective is to provide fact-based analysis on the most common circumstances and types of road traffic crashes leading to serious injuries of MAIS3+ severity. More specifically, the study is directed at providing an understanding of the main circumstances and factors that affect the emergence of serious road traffic injuries, medically coded as MAIS3+, for the following road traffic modes in the EU: pedestrians, bicyclists, motorcyclists and car occupants.

A distance-based variety of nonlinear multivariate data analysis, including weights for objects and variables

In the distance approach to nonlinear multivariate data analysis the focus is on the optimal representation of the relationships between the objects in the analysis. In this paper two methods are presented for including weights in distance-based nonlinear multivariate data analysis. In the first method, weights are assigned to the objects while the second method is concerned with differential weighting of groups of variables. When each analysis variable defines a group the latter method becomes a variable weighting method. For objects the weights are assumed to be given; for groups of variables they may be given, or estimated. These weighting schemes can also be combined and have several important applications. For example, they make it possible to perform efficient analyses of large data sets, to use the distance-based variety of nonlinear multivariate data analysis as an addition to loglinear analysis of multiway contingency tables, and to do stability studies of the solutions by applying the bootstrap on the objects or the variables in the analysis. These and other applications are discussed, and an efficient algorithm is proposed to minimize the corresponding loss function.