Nous proposons de modéliser les données issues des technologies de séquençage du transcriptome (RNA-Seq) à l'aide de la loi binomiale négative, et nous construisons des modèles de segmentation adaptés à leur étude à différentes échelles biologiques, dans le contexte où ces technologies sont devenues un outil précieux pour l'annotation de génome, l'analyse de l'expression des gènes, et la détection de nouveaux transcrits. Nous développons un algorithme de segmentation rapide pour analyser des séries à l'échelle du chromosome, et nous proposons deux méthodes pour l'estimation du nombre de segments, directement lié au nombre de gènes exprimés dans la cellule, qu'ils soient précédemment annotés ou détectés à cette même occasion. L'objectif d'annotation précise des gènes, et plus particulièrement de comparaison des sites de début et fin de transcription entre individus, nous amène naturellement à nous intéresser à la comparaison des localisations de ruptures dans des séries indépendantes. Nous construisons ainsi dans un cadre de segmentation bayésienne des outils de réponse à nos questions pour lesquels nous sommes capable de fournir des mesures d'incertitude. Nous illustrons nos modèles, tous implémentés dans des packages R, sur des données RNA-Seq provenant d'expériences sur la levure, et montrons par exemple que les frontières des introns sont conservées entre conditions tandis que les débuts et fin de transcriptions sont soumis à l'épissage différentiel. / We propose to model the output of transcriptome sequencing technologies (RNA-Seq) using the negative binomial distribution, as well as build segmentation models suited to their study at different biological scales, in the context of these technologies becoming a valuable tool for genome annotation, gene expression analysis, and new-transcript discovery. We develop a fast segmentation algorithm to analyze whole chromosomes series, and we propose two methods for estimating the number of segments, a key feature related to the number of genes expressed in the cell, should they be identified from previous experiments or discovered at this occasion. Research on precise gene annotation, and in particular comparison of transcription boundaries for individuals, naturally leads us to the statistical comparison of change-points in independent series. To address our questions, we build tools, in a Bayesian segmentation framework, for which we are able to provide uncertainty measures. We illustrate our models, all implemented in R packages, on an RNA-Seq dataset from a study on yeast, and show for instance that the intron boundaries are conserved across conditions while the beginning and end of transcripts are subject to differential splicing.

Vehicles being driven for work purposes represent a large proportion of road collision and deaths in the workplace. These observations mean that people driving for work can impose a large burden on organisations and on society. In addition, previous studies identified a fleet driver effect for which there was greater collision risk for those who drive for work compared to the general driving population, even after controlling for exposure. This accentuates the need for both organisational and government policy makers to take steps to reduce the impact of these collisions. No single intervention has been found to solve issues around work-related road safety therefore a range of initiatives have been directed towards the risks associated with drivers, vehicles, journeys and organisations. Many of the interventions, however, lack robust evidence to support their use. The aim of this thesis is to assess organisational interventions to improve work-related road safety by using econometric models on real-world data. The data represents driving claims made between 2005 and 2012 by employees of a large UK company, with a fleet of approximately 35,000 vehicles. The drivers were employed in a variety of roles such as working in technical positions at customer sites or making sales visits. The company has applied a range of strategies to road safety resulting in annual claim reductions of 7.7% compared to only a 4.5% reduction in collisions nationally. The company s data are used to undertake three studies which focused on driver training, manager training and claim segmentation. Statistical models were employed to investigate the effect of two different driver training courses on the frequency of claims while controlling for other factors. The results indicated that driver training courses significantly reduced both the total number of claims and the claim types targeted by the training. The impacts of the interventions were also adjusted for the effects of non-random driver selection and other safety improvements initiated by the company or other agencies. An important finding of this work was that randomly inflated pre-training events accounted for between a third and a quarter of the observed reduction in claims following training. The second study evaluated the impact of management training on claims using multilevel models which allowed for correlation between observations. The study could not confirm that this training was an effective safety intervention. This null result provides an incentive to re-evaluate the implementation of the scheme. The final study identified homogeneous claim segments using statistical models and the impact of training was evaluated on these segments. Such claims were estimated to be reduced by between 32% to 55% following existing driver training courses. This thesis has helped close important gaps and contributed to knowledge in terms of both intervention methodology and the understanding of the effectiveness of work-related road safety interventions. The results, which are already being applied in the case study organisation, demonstrated that training employees in either safe and fuel efficient driving, or low speed manoeuvring, reduced vehicle insurance claims. Further work is necessary to verify the safety value of manager training including gathering detailed information on interactions between managers and drivers.

Development of the Jordan Performance Appraisal System (JPAS) was completed in 1996. This study examined the factor structure of the classroom observation instrument used in the JPAS. Using observed classroom instructional quality ratings of 1220 elementary teachers of Grades 1-6 in the Jordan School District, this study estimated the factor structure of the data and the rater effect on relevant structural parameters. This study also tested for measurement invariance at the within and between levels across teachers of two grade-level groups (a) lower grades: Grades 1-3 and (b) upper grades: Grades 4-6. Factor structure was estimated using complex exploratory factor analysis (EFA) conducted on a subset of the original data. The analysis provided evidence of a three-factor model for the combined groups. The results of multiple confirmatory factor analyses (CFA) conducted using a different subset of the data cross-validated EFA results. Results from multilevel confirmatory factor analysis (MCFA) indicated the three-factor model fit best at both the within and the between levels, and that the intraclass correlation (ICC) was high (.699), indicating significant rater-level variance. Results from a multilevel multigroup confirmatory factor analysis (MLMG-CFA) indicated that the ICC was not significantly different between groups. Results also indicated configural, metric (weak factorial), and scalar (strong factorial) equivalence between groups. This study provided one of the first examples of how to estimate the impact of cluster-level variables such as rater on grouping variables nested at the within level. It provided an example of how to conduct a multilevel multigroup analysis on count data. It also disproved the assumption that counting classroom teaching behaviors was less subjective than using a categorical rating scale. These results will provide substantial information for future developments made to the classroom observation instrument used in the JPAS.

This study examined differences in sexual risk behaviors by gender and over time among 1281 patients (777 males and 504 females) from 12 community-based substance use disorder treatment programs throughout the United States participating in CTN-0032, a randomized control trial conducted within the National Drug Abuse Treatment Clinical Trials Network. Zero-inflated negative binomial and negative binomial models were used in the statistical analysis. Results indicated significant reductions in most types of sexual risk behaviors among substance users regardless of the intervention arms. There were also significant gender differences in sexual risk behaviors. Men (compared with women) reported more condomless sex acts with their non-primary partners (IRR = 1.80, 95 % CI 1.21–2.69) and condomless anal sex acts (IRR = 1.74, 95 % CI 1.11–2.72), but fewer condomless sex partners (IRR = 0.87, 95 % CI 0.77–0.99), condomless vaginal sex acts (IRR = 0.83, 95 % CI 0.69–1.00), and condomless sex acts within 2 h of using drugs or alcohol (IRR = 0.70, 95 % CI 0.53–0.90). Gender-specific intervention approaches are called for in substance use disorder treatment.

No description available.

No description available.

This thesis develops and evaluates a probabilistic model that estimates the stoppage time in football. Stoppage time represents the additional minutes of play given after the original matchtime is over. It is crucial in determining the course of events during the remainder of a match, thereby affecting the odds of live sports betting. The proposed approach uses the negative binomial distribution to model events in football and stoppage time is viewed as a convolution of these distributions. The parameters of the negative binomials are estimated using machine learning methods in Python, with TensorFlow as the underlying framework. The data used for the analysis consists of event data for thousands of football matches with corresponding stoppage time, as well as the duration of pauses that have occurred in these games. The negative binomial distribution is shown to be a good fit and can be adapted to the data using scaling and resolution techniques. The model allows us to see how different events contribute to the stoppage time, and the results indicate that injuries, VAR checks, and red cards have the most significant impact on stoppage time. The model has potential for use in live sports betting and can enhance the accuracy of odds calculation. This work was conducted in collaboration with xAlgo which is a department of Kambi, a business-to-business provider of sports betting services.

No description available.

The objective of this project is to fit a sequence of increasingly complex zero-inflated censored regression models to a known data set. It is quite common to find censored count data in statistical analyses of health-related data. Modeling such data while ignoring the censoring, zero-inflation, and overdispersion often results in biased parameter estimates. This project develops various regression models that can be used to predict a count response variable that is affected by various predictor variables. The regression parameters are estimated with Bayesian analysis using a Markov chain Monte Carlo (MCMC) algorithm. The tests for model adequacy are discussed and the models are applied to an observed data set.

This thesis documents the results of the calibration of the Highway Safety Manual (HSM) safety performance function (SPF) for rural two-lane two-way roadway segments in Utah and the development of new SPFs using negative binomial and hierarchical Bayesian modeling techniques. SPFs estimate the safety of a roadway entity, such as a segment or intersection, in terms of number of crashes. The new SPFs were developed for comparison to the calibrated HSM SPF. This research was performed for the Utah Department of Transportation (UDOT).The study area was the state of Utah. Crash data from 2005-2007 on 157 selected study segments provided a 3-year observed crash frequency to obtain a calibration factor for the HSM SPF and develop new SPFs. The calibration factor for the HSM SPF for rural two-lane two-way roads in Utah is 1.16. This indicates that the HSM underpredicts the number of crashes on rural two-lane two-way roads in Utah by sixteen percent. The new SPFs were developed from the same data that were collected for the HSM calibration, with the addition of new data variables that were hypothesized to have a significant effect on crash frequencies. Negative binomial regression was used to develop four new SPFs, and one additional SPF was developed using hierarchical (or full) Bayesian techniques. The empirical Bayes (EB) method can be applied with each negative binomial SPF because the models include an overdispersion parameter used with the EB method. The hierarchical Bayesian technique is a newer, more mathematically-intense method that accounts for high levels of uncertainty often present in crash modeling. Because the hierarchical Bayesian SPF produces a density function of a predicted crash frequency, a comparison of this density function with an observed crash frequency can help identify segments with significant safety concerns. Each SPF has its own strengths and weaknesses, which include its data requirements and predicting capability. This thesis recommends that UDOT use Equation 5-11 (a new negative binomial SPF) for predicting crashes, because it predicts crashes with reasonable accuracy while requiring much less data than other models. The hierarchical Bayesian process should be used for evaluating observed crash frequencies to identify segments that may benefit from roadway safety improvements.