On the significance of omitted variables in intersection crash modeling

Advances in safety research—trying to improve the collective understanding of motor vehicle crash causes and contributing factors—rest upon the pursuit of numerous lines of research inquiry. The research community has focused considerable attention on analytical methods development (negative binomial models, simultaneous equations, etc.), on better experimental designs (before-after studies, comparison sites, etc.), on improving exposure measures, and on model specification improvements (additive terms, non-linear relations, etc.). One might logically seek to know which lines of inquiry might provide the most significant improvements in understanding crash causation and/or prediction. It is the contention of this paper that the exclusion of important variables (causal or surrogate measures of causal variables) cause omitted variable bias in model estimation and is an important and neglected line of inquiry in safety research. In particular, spatially related variables are often difficult to collect and omitted from crash models—but offer significant opportunities to better understand contributing factors and/or causes of crashes. This study examines the role of important variables (other than Average Annual Daily Traffic (AADT)) that are generally omitted from intersection crash prediction models. In addition to the geometric and traffic regulatory information of intersection, the proposed model includes many spatial factors such as local influences of weather, sun glare, proximity to drinking establishments, and proximity to schools—representing a mix of potential environmental and human factors that are theoretically important, but rarely used. Results suggest that these variables in addition to AADT have significant explanatory power, and their exclusion leads to omitted variable bias. Provided is evidence that variable exclusion overstates the effect of minor road AADT by as much as 40% and major road AADT by 14%.

Severity of driver injury and vehicle damage in traffic crashes at intersections: A Bayesian hierarchical analysis

Most crash severity studies ignored severity correlations between driver-vehicle units involved in the same crashes. Models without accounting for these within-crash correlations will result in biased estimates in the factor effects. This study developed a Bayesian hierarchical binomial logistic model to identify the significant factors affecting the severity level of driver injury and vehicle damage in traffic crashes at signalized intersections. Crash data in Singapore were employed to calibrate the model. Model fitness assessment and comparison using Intra-class Correlation Coefficient (ICC) and Deviance Information Criterion (DIC) ensured the suitability of introducing the crash-level random effects. Crashes occurring in peak time, in good street lighting condition, involving pedestrian injuries are associated with a lower severity, while those in night time, at T/Y type intersections, on right-most lane, and installed with red light camera have larger odds of being severe. Moreover, heavy vehicles have a better resistance on severe crash, while crashes involving two-wheel vehicles, young or aged drivers, and the involvement of offending party are more likely to result in severe injuries.

Modeling random effect and excess zeros in road traffic accident prediction

Poisson distribution has often been used for count like accident data. Negative Binomial (NB) distribution has been adopted in the count data to take care of the over-dispersion problem. However, Poisson and NB distributions are incapable of taking into account some unobserved heterogeneities due to spatial and temporal effects of accident data. To overcome this problem, Random Effect models have been developed. Again another challenge with existing traffic accident prediction models is the distribution of excess zero accident observations in some accident data. Although Zero-Inflated Poisson (ZIP) model is capable of handling the dual-state system in accident data with excess zero observations, it does not accommodate the within-location correlation and between-location correlation heterogeneities which are the basic motivations for the need of the Random Effect models. This paper proposes an effective way of fitting ZIP model with location specific random effects and for model calibration and assessment the Bayesian analysis is recommended.

Beef carcasses with larger eye muscle areas, lower ossification scores and improved nutrition have a lower incidence of dark cutting

This study evaluated the effect of eye muscle area (EMA), ossification, carcass weight, marbling and rib fat depth on the incidence of dark cutting (pH u > 5.7) using routinely collected Meat Standards Australia (MSA) data. Data was obtained from 204,072 carcasses at a Western Australian processor between 2002 and 2008. Binomial data of pH u compliance was analysed using a logit model in a Bayesian framework. Increasing eye muscle area from 40 to 80 cm 2, increased pH u compliance by around 14% (P < 0.001) in carcasses less than 350 kg. As carcass weight increased from 150 kg to 220 kg, compliance increased by 13% (P < 0.001) and younger cattle with lower ossification were also 7% more compliant (P < 0.001). As rib fat depth increased from 0 to 20 mm, pH u compliance increased by around 10% (P < 0.001) yet marbling had no effect on dark cutting. Increasing musculature and growth combined with good nutrition will minimise dark cutting beef in Australia.

Analysis of involved parties in port water conflicts

With increasing rate of shipping traffic, the risk of collisions in busy and congested port waters is expected to rise. However, due to low collision frequencies it is difficult to analyze such risk in a sound statistical manner. This study aims at examining the occurrence of traffic conflicts in order to understand the characteristics of vessels involved in navigational hazards. A binomial logit model was employed to evaluate the association of vessel attributes and the kinematic conditions with conflict severity levels. Results show a positive association for vessels of small gross tonnage, overall vessel length, vessel height and draft with conflict risk. Conflicts involving a pair of dynamic vessels sailing at low speeds also have similar effects.

Statistical analysis of conflict involvements in port water navigation

With increasing rate of shipping traffic, the risk of collisions in busy and congested port waters is likely to rise. However, due to low collision frequencies in port waters, it is difficult to analyze such risk in a sound statistical manner. A convenient approach of investigating navigational collision risk is the application of the traffic conflict techniques, which have potential to overcome the difficulty of obtaining statistical soundness. This study aims at examining port water conflicts in order to understand the characteristics of collision risk with regard to vessels involved, conflict locations, traffic and kinematic conditions. A hierarchical binomial logit model, which considers the potential correlations between observation-units, i.e., vessels, involved in the same conflicts, is employed to evaluate the association of explanatory variables with conflict severity levels. Results show higher likelihood of serious conflicts for vessels of small gross tonnage or small overall length. The probability of serious conflict also increases at locations where vessels have more varied headings, such as traffic intersections and anchorages; becoming more critical at night time. Findings from this research should assist both navigators operating in port waters as well as port authorities overseeing navigational management.

Analytical solutions of the multi-term modified power law wave equations in a finite domain

Fractional partial differential equations with more than one fractional derivative term in time, such as the Szabo wave equation, or the power law wave equation, describe important physical phenomena. However, studies of these multi-term time-space or time fractional wave equations are still under development. In this paper, multi-term modified power law wave equations in a finite domain are considered. The multi-term time fractional derivatives are defined in the Caputo sense, whose orders belong to the intervals (1, 2], [2, 3), [2, 4) or (0, n) (n > 2), respectively. Analytical solutions of the multi-term modified power law wave equations are derived. These new techniques are based on Luchko’s Theorem, a spectral representation of the Laplacian operator, a method of separating variables and fractional derivative techniques. Then these general methods are applied to the special cases of the Szabo wave equation and the power law wave equation. These methods and techniques can also be extended to other kinds of the multi term time-space fractional models including fractional Laplacian.

Analytical solutions for the multi-term time-space Caputo-Riesz fractional advection-diffusion equations on a finite domain

Generalized fractional partial differential equations have now found wide application for describing important physical phenomena, such as subdiffusive and superdiffusive processes. However, studies of generalized multi-term time and space fractional partial differential equations are still under development. In this paper, the multi-term time-space Caputo-Riesz fractional advection diffusion equations (MT-TSCR-FADE) with Dirichlet nonhomogeneous boundary conditions are considered. The multi-term time-fractional derivatives are defined in the Caputo sense, whose orders belong to the intervals [0, 1], [1, 2] and [0, 2], respectively. These are called respectively the multi-term time-fractional diffusion terms, the multi-term time-fractional wave terms and the multi-term time-fractional mixed diffusion-wave terms. The space fractional derivatives are defined as Riesz fractional derivatives. Analytical solutions of three types of the MT-TSCR-FADE are derived with Dirichlet boundary conditions. By using Luchko's Theorem (Acta Math. Vietnam., 1999), we proposed some new techniques, such as a spectral representation of the fractional Laplacian operator and the equivalent relationship between fractional Laplacian operator and Riesz fractional derivative, that enabled the derivation of the analytical solutions for the multi-term time-space Caputo-Riesz fractional advection-diffusion equations. © 2012.

A proposal to change existing virus species names to non-Latinized binomials

A proposal has been posted on the ICTV website (2011. 001aG. N. v1. binomial_sp_names) to replace virus species names by non-Latinized binomial names consisting of the current italicized species name with the terminal word "virus" replaced by the italicized and non-capitalized genus name to which the species belongs. If implemented, the current italicized species name Measles virus, for instance, would become Measles morbillivirus while the current virus name measles virus and its abbreviation MeV would remain unchanged. The rationale for the proposed change is presented. © 2010 Springer-Verlag.

On the commonly accepted assumptions regarding observed motor vehicle crash counts at transport system locations

Readily accepted knowledge regarding crash causation is consistently omitted from efforts to model and subsequently understand motor vehicle crash occurrence and their contributing factors. For instance, distracted and impaired driving accounts for a significant proportion of crash occurrence, yet is rarely modeled explicitly. In addition, spatially allocated influences such as local law enforcement efforts, proximity to bars and schools, and roadside chronic distractions (advertising, pedestrians, etc.) play a role in contributing to crash occurrence and yet are routinely absent from crash models. By and large, these well-established omitted effects are simply assumed to contribute to model error, with predominant focus on modeling the engineering and operational effects of transportation facilities (e.g. AADT, number of lanes, speed limits, width of lanes, etc.) The typical analytical approach—with a variety of statistical enhancements—has been to model crashes that occur at system locations as negative binomial (NB) distributed events that arise from a singular, underlying crash generating process. These models and their statistical kin dominate the literature; however, it is argued in this paper that these models fail to capture the underlying complexity of motor vehicle crash causes, and thus thwart deeper insights regarding crash causation and prevention. This paper first describes hypothetical scenarios that collectively illustrate why current models mislead highway safety researchers and engineers. It is argued that current model shortcomings are significant, and will lead to poor decision-making. Exploiting our current state of knowledge of crash causation, crash counts are postulated to arise from three processes: observed network features, unobserved spatial effects, and ‘apparent’ random influences that reflect largely behavioral influences of drivers. It is argued; furthermore, that these three processes in theory can be modeled separately to gain deeper insight into crash causes, and that the model represents a more realistic depiction of reality than the state of practice NB regression. An admittedly imperfect empirical model that mixes three independent crash occurrence processes is shown to outperform the classical NB model. The questioning of current modeling assumptions and implications of the latent mixture model to current practice are the most important contributions of this paper, with an initial but rather vulnerable attempt to model the latent mixtures as a secondary contribution.

Extremely cold and hot temperatures increase the risk of ischaemic heart disease mortality : epidemiological evidence from China

Objective: To examine the effects of extremely cold and hot temperatures on ischaemic heart disease (IHD) mortality in five cities (Beijing, Tianjin, Shanghai, Wuhan and Guangzhou) in China; and to examine the time relationships between cold and hot temperatures and IHD mortality for each city. Design: A negative binomial regression model combined with a distributed lag non-linear model was used to examine city-specific temperature effects on IHD mortality up to 20 lag days. A meta-analysis was used to pool the cold effects and hot effects across the five cities. Patients: 16 559 IHD deaths were monitored by a sentinel surveillance system in five cities during 2004–2008. Results: The relationships between temperature and IHD mortality were non-linear in all five cities. The minimum-mortality temperatures in northern cities were lower than in southern cities. In Beijing, Tianjin and Guangzhou, the effects of extremely cold temperatures were delayed, while Shanghai and Wuhan had immediate cold effects. The effects of extremely hot temperatures appeared immediately in all the cities except Wuhan. Meta-analysis showed that IHD mortality increased 48% at the 1st percentile of temperature (extremely cold temperature) compared with the 10th percentile, while IHD mortality increased 18% at the 99th percentile of temperature (extremely hot temperature) compared with the 90th percentile. Conclusions: Results indicate that both extremely cold and hot temperatures increase IHD mortality in China. Each city has its characteristics of heat effects on IHD mortality. The policy for response to climate change should consider local climate–IHD mortality relationships.

L’internationalisation des jeunes entreprises de hautes technologies : Le rôle des compétences internationales des fondateurs [The internationalization of new technology based firms : the role of the international skills of founding teams]

Extending recent research on the importance of specific resources and skills for the internationalization of start-ups, this article tests a negative binomial model on a sample of 520 recently created high technology firms from the UK and Germany. The results show that previous international experience of entrepreneurs facilitates the rapid penetration of foreign markets, especially when the company features a clear and deliberate strategic intent of internationalization from the outset. This research provides one of the first empirical studies linking the influence of entrepreneurial teams to a high probability of success in the internationalization of high-technology ventures.

Factors associated with common mental disorders during the transition to adulthood : do emerging adults and ‘early starters’ differ? Findings from an Australian birth cohort study

Background: Despite increasing diversity in pathways to adulthood, choices available to young people are influenced by environmental, familial and individual factors, namely access to socioeconomic resources, family support and mental and physical health status. Young people from families with higher socioeconomic position (SEP) are more likely to pursue tertiary education and delay entry to adulthood, whereas those from low socioeconomic backgrounds are less likely to attain higher education or training, and more likely to partner and become parents early. The first group are commonly termed ‘emerging adults’ and the latter group ‘early starters’. Mental health disorders during this transition can seriously disrupt psychological, social and academic development as well as employment prospects. Depression, anxiety and most substance use disorders have early onset during adolescence and early adulthood with approximately three quarters of lifetime psychiatric disorders having emerged by 24 years of age. Aims: This thesis aimed to explore the relationships between mental health, sociodemographic factors and family functioning during the transition to adulthood. Four areas were investigated: 1) The key differences between emerging adults and ‘early starters’, were examined and focused on a series of social, economic, and demographic factors as well as DSM-IV diagnoses; 2) Methodological issues associated with the measurement of depression and anxiety in young adults were explored by comparing a quantitative measure of symptoms of anxiety and depression (Achenbach’s YSR and YASR internalising scales) with DSM-IV diagnosed depression and anxiety. 3) The association between family SEP and DSM-IV depression and anxiety was examined in relation to the different pathways to adulthood. 4) Finally, the association between pregnancy loss, abortion and miscarriage, and DSM-IV diagnoses of common psychiatric disorders was assessed in young women who reported early parenting, experiencing a pregnancy loss, or who had never been pregnant. Methods: Data were taken from the Mater University Study of Pregnancy (MUSP), a large birth cohort started in 1981 in Brisbane, Australia. 7223 mothers and their children were assessed five times, at 6 months, 5, 14 and 21 years after birth. Over 3700 young adults, aged 18 to 23 years, were interviewed at the 21-year phase. Respondents completed an extensive series of self-reported questionnaires and a computerised structured psychiatric interview. Three outcomes were assessed at the 21-year phase. Mental health disorders diagnosed by a computerised structured psychiatric interview (CIDI-Auto), the prevalence of DSM-IV depression, anxiety and substance use disorders within the previous 12-month, during the transition (between ages of 18 and 23 years) or lifetime were examined. The primary outcome “current stage in the transition to adulthood” was developed using a measure conceptually constructed from the literature. The measure was based on important demographic markers, and these defined four independent groups: emerging adults (single with no children and living with parents), and three categories of ‘early starter’, singles (with no children or partner, living independently), those with a partner (married or cohabitating but without children) and parents. Early pregnancy loss was assessed using a measure that also defined four independent groups and was based on pregnancy outcomes in the young women This categorised the young women into those who were never pregnant, women who gave birth to a live child, and women who reported some form of pregnancy loss, either an abortion or a spontaneous miscarriage. A series of analyses were undertaken to test the study aims. Potential confounding and mediating factors were prospectively measured between the child’s birth and the 21-year phase. Binomial and multinomial logistic regression was used to estimate the risk of relevant outcomes, and the associations were reported as odds ratios (OR) and 95% confidence intervals (95%CI). Key findings: The thesis makes a number of important contributions to our understanding of the transition to adulthood, particularly in relation to the mental health consequences associated with different pathways. Firstly, findings from the thesis clearly showed that young people who parented or partnered early fared worse across most of the economic and social factors as well as the common mental disorders when compared to emerging adults. That is, young people who became early parents were also more likely to experience recent anxiety (OR=2.0, 95%CI 1.5-2.8) and depression (OR=1.7, 95%CI 1.1-2.7) than were emerging adults after taking into account a range of confounding factors. Singles and those partnering early also had higher rates of lifetime anxiety and depression than emerging adults. Young people who partnered early, but were without children, had decreased odds of recent depression; this may be due to the protective effect of early marriage against depression. It was also found that young people who form families early had an increased risk of cigarette smoking (parents OR=3.7, 95%CI 2.9-4.8) compared to emerging adults, but not heavy alcohol (parents OR=0.4, 95%CI 0.3-0.6) or recent illicit drug use. The high rates of cigarette smoking and tobacco use disorders in ‘early starters’ were explained by common risk factors related to early adversity and lower SEP. Having a child and early marriage may well function as a ‘turning point’ for some young people, it is not clear whether this is due to a conscious decision to disengage from a previous ‘substance using’ lifestyle or simply that they no longer have the time to devote to such activities because of child caring. In relation to the methodological issues associated with assessing common mental disorders in young adults, it was found that although the Achenbach empirical internalising scales successfully predicted both later DSM-IV depression (YSR OR=2.3, 95%CI 1.7-3.1) and concurrently diagnosed depression (YASR OR=6.9, 95%CI 5.0- 9.5) and anxiety (YASR OR=5.1, 95%CI 3.8- 6.7), the scales discriminated poorly between young people with or without DSM-IV diagnosed mood disorder. Sensitivity values (the proportion of true positives) for the internalising scales were surprisingly low. Only a third of young people with current DSM-IV depression (range for each of the scales was between 34% to 42%) were correctly identified as cases by the YASR internalising scales, and only a quarter with current anxiety disorder (range of 23% to 31%) were correctly identified. Also, use of the DSM-oriented scales increased sensitivity only marginally (for depression between 2-8%, and anxiety between 2-6%) above the standard Achenbach scales. This is despite the fact that the DSM-oriented scales were originally developed to overcome the poor prediction of DSM-IV diagnoses by the Achenbach scales. The internalising scales, both standard and DSM-oriented, were much more effective at identifying young people with comorbid depression and anxiety, with OR’s 10.1 to 21.7 depending on the internalising scale used. SEP is an important predictor of both an early transition to adulthood and the experience of anxiety during that time Family income during adolescence was a strong predictor of early parenting and partnering before age 24 but not early independent living. Compared to families in the upper quintile, young people from families with low income were nearly twice as likely to live with a partner and four times more likely to become parents (OR ranged from 2.6 to 4.0). This association remained after adjusting for current employment and education level. Children raised in low income families were 30% more likely to have an anxiety disorder (OR=1.3, 95%CI 0.9-1.9), but not depression, as young adults when compared to children from wealthier families. Emerging adults and ‘early starters’ from low income families did not differ in their likelihood of having a later anxiety disorder. Young women reporting a pregnancy loss had nearly three times the odds of experiencing a lifetime illicit drug disorder (excluding cannabis) [abortion OR=3.6, 95%CI 2.0-6.7 and miscarriage OR=2.6, 95%CI 1.2-5.4]. Abortion was associated with alcohol use disorder (OR=2.1, 95%CI 1.3- 3.5) and 12-month depression (OR=1.9, 95%CI 1.1- 3.1). These finding suggest that the association identified by Fergusson et al between abortion and later psychiatric disorders in young women may be due to pregnancy loss and not to abortion, per se. Conclusion: Findings from this thesis support the view that young people who parent or partner early have a greater burden of depression and anxiety when compared to emerging adults. As well, young women experiencing pregnancy loss, from either abortion or miscarriage, are more likely to experience depression and anxiety than are those who give birth to a live infant or who have never been pregnant. Depression, anxiety and substance use disorders often go unrecognised and untreated in young people; this is especially true in young people from lower SEP. Early identification of these common mental health disorders is important, as depression and anxiety experienced during the transition to adulthood have been found to seriously disrupt an individual’s social, educational and economic prospects in later life.

Sampling stored-product insect pests : a comparison of four statistical sampling models for probability of pest detection

Background: Developing sampling strategies to target biological pests such as insects in stored grain is inherently difficult owing to species biology and behavioural characteristics. The design of robust sampling programmes should be based on an underlying statistical distribution that is sufficiently flexible to capture variations in the spatial distribution of the target species. Results: Comparisons are made of the accuracy of four probability-of-detection sampling models - the negative binomial model,1 the Poisson model,1 the double logarithmic model2 and the compound model3 - for detection of insects over a broad range of insect densities. Although the double log and negative binomial models performed well under specific conditions, it is shown that, of the four models examined, the compound model performed the best over a broad range of insect spatial distributions and densities. In particular, this model predicted well the number of samples required when insect density was high and clumped within experimental storages. Conclusions: This paper reinforces the need for effective sampling programs designed to detect insects over a broad range of spatial distributions. The compound model is robust over a broad range of insect densities and leads to substantial improvement in detection probabilities within highly variable systems such as grain storage.