Extension of the Application of Conway‐Maxwell‐Poisson Models: Analyzing Traffic Crash Data Exhibiting Underdispersion

The objective of this article is to evaluate the performance of the COM‐Poisson GLM for analyzing crash data exhibiting underdispersion (when conditional on the mean). The COM‐Poisson distribution, originally developed in 1962, has recently been reintroduced by statisticians for analyzing count data subjected to either over‐ or underdispersion. Over the last year, the COM‐Poisson GLM has been evaluated in the context of crash data analysis and it has been shown that the model performs as well as the Poisson‐gamma model for crash data exhibiting overdispersion. To accomplish the objective of this study, several COM‐Poisson models were estimated using crash data collected at 162 railway‐highway crossings in South Korea between 1998 and 2002. This data set has been shown to exhibit underdispersion when models linking crash data to various explanatory variables are estimated. The modeling results were compared to those produced from the Poisson and gamma probability models documented in a previous published study. The results of this research show that the COM‐Poisson GLM can handle crash data when the modeling output shows signs of underdispersion. Finally, they also show that the model proposed in this study provides better statistical performance than the gamma probability and the traditional Poisson models, at least for this data set.     

Validation of Multimedia Models Assessing Exposure to PAHs—The SOLEX Study

Polluted soils have become a public health problem. While population exposure to soil pollutants is generally quantified using multimedia models, their estimations have not been validated, and studies that attempted to do so are scarce. The objective of the SOLEX study was to compare the predictions of pyrene exposure levels (converted into 1 hydroxypyrene) computed by several models with the results of urinary 1-hydropyrene (1-HOP) assays among 110 employees working at three sites polluted during their past use as manufactured gas plants. Four models were used: AERIS (Canada), CalTOX (California, USA), CLEA (UK), and HESP (The Netherlands). Three occupational exposure scenarios--with office, mixed, and outdoor workers--were constructed, based upon job activities during two measurement campaigns, one in winter and one in summer. The exposure levels estimated by the four models could differ markedly (from 7 up to 80 times) according to the exposure scenario. Also, the predominant exposure routes differed according to the model (direct soil ingestion for HESP and CalTOX, inhalation for AERIS, and dermal absorption for CLEA). The predictions of CalTOX are consistent with the 1-HOP measurements for all the scenarios. For HESP, the consistency is observed for the scenarios, office and mixed, for which the pyrene level in the soil is low. AERIS and CLEA yield results that are systematically above the 1-HOP measurements. This study confirms that validation of the models is crucial and points out to the need to proceed to assess components of the models that are the most influential using appropriate statistical analysis in combination with true field data.     

Wavelet‐Based Testing for Serial Correlation of Unknown Form in Panel Models

Wavelet analysis is a new mathematical method developed as a unified field of science over the last decade or so. As a spatially adaptive analytic tool, wavelets are useful for capturing serial correlation where the spectrum has peaks or kinks, as can arise from persistent dependence, seasonality, and other kinds of periodicity. This paper proposes a new class of generally applicable wavelet‐based tests for serial correlation of unknown form in the estimated residuals of a panel regression model, where error components can be one‐way or two‐way, individual and time effects can be fixed or random, and regressors may contain lagged dependent variables or deterministic/stochastic trending variables. Our tests are applicable to unbalanced heterogenous panel data. They have a convenient null limit N(0,1) distribution. No formulation of an alternative model is required, and our tests are consistent against serial correlation of unknown form even in the presence of substantial inhomogeneity in serial correlation across individuals. This is in contrast to existing serial correlation tests for panel models, which ignore inhomogeneity in serial correlation across individuals by assuming a common alternative, and thus have no power against the alternatives where the average of serial correlations among individuals is close to zero. We propose and justify a data‐driven method to choose the smoothing parameter—the finest scale in wavelet spectral estimation, making the tests completely operational in practice. The data‐driven finest scale automatically converges to zero under the null hypothesis of no serial correlation and diverges to infinity as the sample size increases under the alternative, ensuring the consistency of our tests. Simulation shows that our tests perform well in small and finite samples relative to some existing tests.     

Simulation of Longitudinal Exposure Data with Variance‐Covariance Structures Based on Mixed Models

Longitudinal data are important in exposure and risk assessments, especially for pollutants with long half‐lives in the human body and where chronic exposures to current levels in the environment raise concerns for human health effects. It is usually difficult and expensive to obtain large longitudinal data sets for human exposure studies. This article reports a new simulation method to generate longitudinal data with flexible numbers of subjects and days. Mixed models are used to describe the variance‐covariance structures of input longitudinal data. Based on estimated model parameters, simulation data are generated with similar statistical characteristics compared to the input data. Three criteria are used to determine similarity: the overall mean and standard deviation, the variance components percentages, and the average autocorrelation coefficients. Upon the discussion of mixed models, a simulation procedure is produced and numerical results are shown through one human exposure study. Simulations of three sets of exposure data successfully meet above criteria. In particular, simulations can always retain correct weights of inter‐ and intrasubject variances as in the input data. Autocorrelations are also well followed. Compared with other simulation algorithms, this new method stores more information about the input overall distribution so as to satisfy the above multiple criteria for statistical targets. In addition, it generates values from numerous data sources and simulates continuous observed variables better than current data methods. This new method also provides flexible options in both modeling and simulation procedures according to various user requirements.     

Improved Transferability of Data‐Driven Damage Models Through Sample Selection Bias Correction

Damage models for natural hazards are used for decision making on reducing and transferring risk. The damage estimates from these models depend on many variables and their complex sometimes nonlinear relationships with the damage. In recent years, data‐driven modeling techniques have been used to capture those relationships. The available data to build such models are often limited. Therefore, in practice it is usually necessary to transfer models to a different context. In this article, we show that this implies the samples used to build the model are often not fully representative for the situation where they need to be applied on, which leads to a “sample selection bias.” In this article, we enhance data‐driven damage models by applying methods, not previously applied to damage modeling, to correct for this bias before the machine learning (ML) models are trained. We demonstrate this with case studies on flooding in Europe, and typhoon wind damage in the Philippines. Two sample selection bias correction methods from the ML literature are applied and one of these methods is also adjusted to our problem. These three methods are combined with stochastic generation of synthetic damage data. We demonstrate that for both case studies, the sample selection bias correction techniques reduce model errors, especially for the mean bias error this reduction can be larger than 30%. The novel combination with stochastic data generation seems to enhance these techniques. This shows that sample selection bias correction methods are beneficial for damage model transfer.     

The Dynamics of Pre‐ and Post‐purchase Service and Consumer Evaluation of Online Retailers: A Comparative Analysis of Dissonance and Disconfirmation Models*

This article explores the theoretical underpinnings of the dissonance framework in online consumer satisfaction formation process. Specifically, we suggest that any discrepancy between pre‐ and post‐purchase service performance would help determine consumers’ evaluations of online vendors. Drawing upon cognitive dissonance theory, a conceptual model is developed and tested in two different studies (preliminary and main studies). Using data from 191 college students collected longitudinally, the preliminary study demonstrates the validity and reliability of the measurements. Using a comparative analysis, the main study then tests our conceptual model as well as various competing models, including the expectation–confirmation model, with a sample of 292 online consumers. The results in both studies support our main prediction that the service encountered in different stages establishes dissonance. Specifically, we find that dissonance explains online consumers’ satisfaction process to a substantial extent, as compared with disconfirmation under the same conditions in online retailers. This study contributes to providing an alternative yet substantial approach for expectation–confirmation theory, reflecting the overarching nature of online shopping.     

Asymptotic Distributions of Quasi‐Maximum Likelihood Estimators for Spatial Autoregressive Models

This paper investigates asymptotic properties of the maximum likelihood estimator and the quasi‐maximum likelihood estimator for the spatial autoregressive model. The rates of convergence of those estimators may depend on some general features of the spatial weights matrix of the model. It is important to make the distinction with different spatial scenarios. Under the scenario that each unit will be influenced by only a few neighboring units, the estimators may have ‐rate of convergence and be asymptotically normal. When each unit can be influenced by many neighbors, irregularity of the information matrix may occur and various components of the estimators may have different rates of convergence.     

QMRA for Drinking Water: 1. Revisiting the Mathematical Structure of Single‐Hit Dose‐Response Models

Dose‐response models are essential to quantitative microbial risk assessment (QMRA), providing a link between levels of human exposure to pathogens and the probability of negative health outcomes. In drinking water studies, the class of semi‐mechanistic models known as single‐hit models, such as the exponential and the exact beta‐Poisson, has seen widespread use. In this work, an attempt is made to carefully develop the general mathematical single‐hit framework while explicitly accounting for variation in (1) host susceptibility and (2) pathogen infectivity. This allows a precise interpretation of the so‐called single‐hit probability and precise identification of a set of statistical independence assumptions that are sufficient to arrive at single‐hit models. Further analysis of the model framework is facilitated by formulating the single‐hit models compactly using probability generating and moment generating functions. Among the more practically relevant conclusions drawn are: (1) for any dose distribution, variation in host susceptibility always reduces the single‐hit risk compared to a constant host susceptibility (assuming equal mean susceptibilities), (2) the model‐consistent representation of complete host immunity is formally demonstrated to be a simple scaling of the response, (3) the model‐consistent expression for the total risk from repeated exposures deviates (gives lower risk) from the conventional expression used in applications, and (4) a model‐consistent expression for the mean per‐exposure dose that produces the correct total risk from repeated exposures is developed.     

QMRA for Drinking Water: 2. The Effect of Pathogen Clustering in Single‐Hit Dose‐Response Models

Spatial and/or temporal clustering of pathogens will invalidate the commonly used assumption of Poisson‐distributed pathogen counts (doses) in quantitative microbial risk assessment. In this work, the theoretically predicted effect of spatial clustering in conventional “single‐hit” dose‐response models is investigated by employing the stuttering Poisson distribution, a very general family of count distributions that naturally models pathogen clustering and contains the Poisson and negative binomial distributions as special cases. The analysis is facilitated by formulating the dose‐response models in terms of probability generating functions. It is shown formally that the theoretical single‐hit risk obtained with a stuttering Poisson distribution is lower than that obtained with a Poisson distribution, assuming identical mean doses. A similar result holds for mixed Poisson distributions. Numerical examples indicate that the theoretical single‐hit risk is fairly insensitive to moderate clustering, though the effect tends to be more pronounced for low mean doses. Furthermore, using Jensen's inequality, an upper bound on risk is derived that tends to better approximate the exact theoretical single‐hit risk for highly overdispersed dose distributions. The bound holds with any dose distribution (characterized by its mean and zero inflation index) and any conditional dose‐response model that is concave in the dose variable. Its application is exemplified with published data from Norovirus feeding trials, for which some of the administered doses were prepared from an inoculum of aggregated viruses. The potential implications of clustering for dose‐response assessment as well as practical risk characterization are discussed.     

Understanding the Role of Individual Innovativeness in the Acceptance of IT‐Based Innovations: Comparative Analyses of Models and Measures*

As new technological innovations are rapidly introduced and changed, identifying an individual characteristic that has a persistent effect on the acceptance decisions across multiple technologies is of substantial value for the successful implementation of information systems. Augmenting prior work on individual innovativeness within the context of information technology, we developed a new measure of adopter category innovativeness (ACI) and compared its effectiveness with the existing measure of personal innovativeness in IT (PIIT). Further, we examined two alternative models in which the role of individual innovativeness was theorized differently—either as a moderator of the effects the perceived innovation characteristics of usefulness, ease of use, and compatibility have on future use intention (moderator model) or as a direct determinant of the innovation characteristics (direct determinant model). To ensure the generalizability of the study findings, two field studies (N= 634) were conducted, each of which examined the two models (moderator and direct determinant) and measured individual innovativeness using the two measures (ACI and PIIT). Study 1 surveyed the online buying practices of 412 individuals, and Study 2 surveyed personal digital assistant adoption of 222 healthcare professionals. Across the markedly different adoption contexts, the study results consistently show that individual innovativeness is a direct determinant of the innovation characteristics, and the two measures share many commonalities. The new measure offers some additional utilities not found in the PIIT measure by allowing individuals to be directly classified and mapped into adopter categories. Implications are drawn for future research and practice.     

Absolute Risk or Relative Risk? A Study of Intraspecies and Interspecies Extrapolation of Chemical‐Induced Cancer Risk

We have used the CBDS database of the National Toxicology Program to study the difference between absolute risk and relative risk models for interspecies and intersex predictions of cancer risk. For no combination (class) of tumor and site is the prediction good for all chemicals. The variation in predicted risk between chemicals exceeds the difference in risks resulting from application of these two models. On the whole, it appears that relative risk is a better model.     

Towards a Fuzzy Bayesian Network Based Approach for Safety Risk Analysis of Tunnel‐Induced Pipeline Damage

Tunneling excavation is bound to produce significant disturbances to surrounding environments, and the tunnel‐induced damage to adjacent underground buried pipelines is of considerable importance for geotechnical practice. A fuzzy Bayesian networks (FBNs) based approach for safety risk analysis is developed in this article with detailed step‐by‐step procedures, consisting of risk mechanism analysis, the FBN model establishment, fuzzification, FBN‐based inference, defuzzification, and decision making. In accordance with the failure mechanism analysis, a tunnel‐induced pipeline damage model is proposed to reveal the cause‐effect relationships between the pipeline damage and its influential variables. In terms of the fuzzification process, an expert confidence indicator is proposed to reveal the reliability of the data when determining the fuzzy probability of occurrence of basic events, with both the judgment ability level and the subjectivity reliability level taken into account. By means of the fuzzy Bayesian inference, the approach proposed in this article is capable of calculating the probability distribution of potential safety risks and identifying the most likely potential causes of accidents under both prior knowledge and given evidence circumstances. A case concerning the safety analysis of underground buried pipelines adjacent to the construction of the Wuhan Yangtze River Tunnel is presented. The results demonstrate the feasibility of the proposed FBN approach and its application potential. The proposed approach can be used as a decision tool to provide support for safety assurance and management in tunnel construction, and thus increase the likelihood of a successful project in a complex project environment.     

A Probabilistic Framework for Risk Analysis of Widespread Flood Events: A Proof‐of‐Concept Study

This article presents a flood risk analysis model that considers the spatially heterogeneous nature of flood events. The basic concept of this approach is to generate a large sample of flood events that can be regarded as temporal extrapolation of flood events. These are combined with cumulative flood impact indicators, such as building damages, to finally derive time series of damages for risk estimation. Therefore, a multivariate modeling procedure that is able to take into account the spatial characteristics of flooding, the regionalization method top‐kriging, and three different impact indicators are combined in a model chain. Eventually, the expected annual flood impact (e.g., expected annual damages) and the flood impact associated with a low probability of occurrence are determined for a study area. The risk model has the potential to augment the understanding of flood risk in a region and thereby contribute to enhanced risk management of, for example, risk analysts and policymakers or insurance companies. The modeling framework was successfully applied in a proof‐of‐concept exercise in Vorarlberg (Austria). The results of the case study show that risk analysis has to be based on spatially heterogeneous flood events in order to estimate flood risk adequately.     

Development of Dose‐Response Models to Predict the Relationship for Human Toxoplasma gondii Infection Associated with Meat Consumption

Toxoplasma gondii is a protozoan parasite that is responsible for approximately 24% of deaths attributed to foodborne pathogens in the United States. It is thought that a substantial portion of human T. gondii infections is acquired through the consumption of meats. The dose‐response relationship for human exposures to T. gondii‐infected meat is unknown because no human data are available. The goal of this study was to develop and validate dose‐response models based on animal studies, and to compute scaling factors so that animal‐derived models can predict T. gondii infection in humans. Relevant studies in literature were collected and appropriate studies were selected based on animal species, stage, genotype of T. gondii, and route of infection. Data were pooled and fitted to four sigmoidal‐shaped mathematical models, and model parameters were estimated using maximum likelihood estimation. Data from a mouse study were selected to develop the dose‐response relationship. Exponential and beta‐Poisson models, which predicted similar responses, were selected as reasonable dose‐response models based on their simplicity, biological plausibility, and goodness fit. A confidence interval of the parameter was determined by constructing 10,000 bootstrap samples. Scaling factors were computed by matching the predicted infection cases with the epidemiological data. Mouse‐derived models were validated against data for the dose‐infection relationship in rats. A human dose‐response model was developed as P (d) = 1–exp (–0.0015 × 0.005 × d) or P (d) = 1–(1 + d × 0.003 / 582.414)^?1.479. Both models predict the human response after consuming T. gondii‐infected meats, and provide an enhanced risk characterization in a quantitative microbial risk assessment model for this pathogen.     

New Data,Strategies, and Insights for Listeria monocytogenes Dose‐Response Models: Summary of an Interagency Workshop, 2011

Listeria monocytogenes is a leading cause of hospitalization, fetal loss, and death due to foodborne illnesses in the United States. A quantitative assessment of the relative risk of listeriosis associated with the consumption of 23 selected categories of ready‐to‐eat foods, published by the U.S. Department of Health and Human Services and the U.S. Department of Agriculture in 2003, has been instrumental in identifying the food products and practices that pose the greatest listeriosis risk and has guided the evaluation of potential intervention strategies. Dose‐response models, which quantify the relationship between an exposure dose and the probability of adverse health outcomes, were essential components of the risk assessment. However, because of data gaps and limitations in the available data and modeling approaches, considerable uncertainty existed. Since publication of the risk assessment, new data have become available for modeling L. monocytogenes dose‐response. At the same time, recent advances in the understanding of L. monocytogenes pathophysiology and strain diversity have warranted a critical reevaluation of the published dose‐response models. To discuss strategies for modeling L. monocytogenes dose‐response, the Interagency Risk Assessment Consortium (IRAC) and the Joint Institute for Food Safety and Applied Nutrition (JIFSAN) held a scientific workshop in 2011 (details available at http://foodrisk.org/irac/events/ ). The main findings of the workshop and the most current and relevant data identified during the workshop are summarized and presented in the context of L. monocytogenes dose‐response. This article also discusses new insights on dose‐response modeling for L. monocytogenes and research opportunities to meet future needs.     

Reanalysis of the DEMS Nested Case‐Control Study of Lung Cancer and Diesel Exhaust: Suitability for Quantitative Risk Assessment

The International Agency for Research on Cancer (IARC) in 2012 upgraded its hazard characterization of diesel engine exhaust (DEE) to “carcinogenic to humans.” The Diesel Exhaust in Miners Study (DEMS) cohort and nested case‐control studies of lung cancer mortality in eight U.S. nonmetal mines were influential in IARC's determination. We conducted a reanalysis of the DEMS case‐control data to evaluate its suitability for quantitative risk assessment (QRA). Our reanalysis used conditional logistic regression and adjusted for cigarette smoking in a manner similar to the original DEMS analysis. However, we included additional estimates of DEE exposure and adjustment for radon exposure. In addition to applying three DEE exposure estimates developed by DEMS, we applied six alternative estimates. Without adjusting for radon, our results were similar to those in the original DEMS analysis: all but one of the nine DEE exposure estimates showed evidence of an association between DEE exposure and lung cancer mortality, with trend slopes differing only by about a factor of two. When exposure to radon was adjusted, the evidence for a DEE effect was greatly diminished, but was still present in some analyses that utilized the three original DEMS DEE exposure estimates. A DEE effect was not observed when the six alternative DEE exposure estimates were utilized and radon was adjusted. No consistent evidence of a DEE effect was found among miners who worked only underground. This article highlights some issues that should be addressed in any use of the DEMS data in developing a QRA for DEE.     

An Exploratory Study of Procurement Strategies for Multi‐Item RFQs in B2B Markets: Antecedents and Impact on Performance

This research explores procurement strategies for multi‐item requests for quotation (RFQs) in business‐to‐business (B2B) markets using responses from 825 purchasing professionals. The study first establishes procurement strategies that differ based on their level of strategic emphasis, i.e., the importance that is placed on the pursuit of four strategic objectives. Underlying objectives, which are obtained via factor analysis, include the focus on price, security of supply, internal procurement efficiencies, and bundle building. Next, cluster analysis is used to derive prototypical strategic approaches. The three cluster groups that emerge possess the same relative ranking of the four objectives, but differ based on the intensity with which these objectives are pursued. The clusters are labelled as the three strategic groups of strategists, opportunists, and responders. The research then explores, using an industrial buyer behavior lens, the impact of environmental antecedents in determining a particular strategy. Environmental variables include purchase importance, market uncertainty, supply base availability, buyer bargaining power, item experience, and supply base experience. Finally, the study tests the impact of procurement strategy on the buyer's perceived performance, suggesting that strategists, placing more emphasis on the pursuit of strategic sourcing objectives, achieve better performance than opportunists and responders.     

A Tiered Framework for Risk‐Relevant Characterization and Ranking of Chemical Exposures: Applications to the National Children's Study (NCS)

A challenge for large‐scale environmental health investigations such as the National Children's Study (NCS), is characterizing exposures to multiple, co‐occurring chemical agents with varying spatiotemporal concentrations and consequences modulated by biochemical, physiological, behavioral, socioeconomic, and environmental factors. Such investigations can benefit from systematic retrieval, analysis, and integration of diverse extant information on both contaminant patterns and exposure‐relevant factors. This requires development, evaluation, and deployment of informatics methods that support flexible access and analysis of multiattribute data across multiple spatiotemporal scales. A new “Tiered Exposure Ranking” (TiER) framework, developed to support various aspects of risk‐relevant exposure characterization, is described here, with examples demonstrating its application to the NCS. TiER utilizes advances in informatics computational methods, extant database content and availability, and integrative environmental/exposure/biological modeling to support both “discovery‐driven” and “hypothesis‐driven” analyses. “Tier 1” applications focus on “exposomic” pattern recognition for extracting information from multidimensional data sets, whereas second and higher tier applications utilize mechanistic models to develop risk‐relevant exposure metrics for populations and individuals. In this article, “tier 1” applications of TiER explore identification of potentially causative associations among risk factors, for prioritizing further studies, by considering publicly available demographic/socioeconomic, behavioral, and environmental data in relation to two health endpoints (preterm birth and low birth weight). A “tier 2” application develops estimates of pollutant mixture inhalation exposure indices for NCS counties, formulated to support risk characterization for these endpoints. Applications of TiER demonstrate the feasibility of developing risk‐relevant exposure characterizations for pollutants using extant environmental and demographic/socioeconomic data.