The study and practice of epidemiology and public health benefit from the use of mortality statistics, such as mortality rates, which are frequently used as key health indicators. Furthermore, multiple causes of death (MCOD) data offer important information that could not possibly be gathered from other mortality data. This study aimed to describe the interrelationships between various causes of death in the United States in order to improve the understanding of the coexistence of MCOD and thereby improve public health and enhance longevity. The social support theory was used as a framework, and multivariate linear regression analyses were conducted to examine the coexistence of MCOD in approximately 80 million death cases across the United States from 1959 to 2005. The findings showed that in the United States, there is a statistically significant relationship between the number of coexisting MCOD, race, education, and the state of residence. Furthermore, age, gender, and marital status statistically influence the average number of coexisting MCOD. The results offer insights into how the number of coexisting MCOD vary across the United States, races, education levels, gender, age, and marital status and lay a foundation for further investigation into what people are dying from. The results have the long-term potential of helping public health practitioners identify individuals or communities that are at higher risks of death from a number of coexisting MCOD such that actions could be taken to lower the risks to improve people's wellbeing, enhance longevity, and contribute to positive social change.

Low profit margins threaten the sustainability of quick service restaurants (QSRs). In the United States, low levels of employee job satisfaction and low employee perceptions of supervisor support decrease organizational profitability by as much as $151 million annually, depending on the size and type of organization. Guided by the 2-factor theory of motivation, the purpose of this correlational study was to examine the relationship between employee job satisfaction, employee perceptions of supervisor support, and organizational profitability. A convenience sample of employees from 86 QSR franchise locations in Houston, Texas completed the Job Satisfaction and Perceived Supervisor Support surveys. Multiple linear regression analysis and Bonferroni corrected significance calculation predicted organizational profitability (F(2, 71) = 9.20, p < .001, R2 = .206) and employee job satisfaction (ï?¢ = .577, p = .025). The effect size indicated that the regression model accounted for approximately 21% of the variance in organizational profitability. Employee perceptions of supervisor support (ï?¢ = -.140, p = .580) did not relate to any significant variation in organizational profitability. The findings may be of value to QSR business professionals developing initiatives to improve organizational profitability. Improving employees' perceptions of supervisor support to generate high levels of employee job satisfaction could affect behavioral social change to enhance the health and wellbeing of employees and the wealth and sustainability of QSR franchise locations.

Cette thèse aborde le sujet de la régression linéaire dans différents cadres, liés notamment à l’apprentissage. Les deux premiers chapitres présentent le contexte des travaux, leurs apports et les outils mathématiques utilisés. Le troisième chapitre est consacré à la construction d’une fonction de régularisation optimale, permettant par exemple d’améliorer sur le plan théorique la régularisation de l’estimateur LASSO. Le quatrième chapitre présente, dans le domaine de l’optimisation convexe séquentielle, des accélérations d’un algorithme récent et prometteur, MetaGrad, et une conversion d’un cadre dit “séquentiel déterministe" vers un cadre dit “batch stochastique" pour cet algorithme. Le cinquième chapitre s’intéresse à des prévisions successives par intervalles, fondées sur l’agrégation de prédicteurs, sans retour d’expérience intermédiaire ni modélisation stochastique. Enfin, le sixième chapitre applique à un jeu de données pétrolières plusieurs méthodes d’agrégation, aboutissant à des prévisions ponctuelles court-terme et des intervalles de prévision long-terme. / This thesis tackles the topic of linear regression, within several frameworks, mainly linked to statistical learning. The first and second chapters present the context, the results and the mathematical tools of the manuscript. In the third chapter, we provide a way of building an optimal regularization function, improving for instance, in a theoretical way, the LASSO estimator. The fourth chapter presents, in the field of online convex optimization, speed-ups for a recent and promising algorithm, MetaGrad, and shows how to transfer its guarantees from a so-called “online deterministic setting" to a “stochastic batch setting". In the fifth chapter, we introduce a new method to forecast successive intervals by aggregating predictors, without intermediate feedback nor stochastic modeling. The sixth chapter applies several aggregation methods to an oil production dataset, forecasting short-term precise values and long-term intervals.

This diploma thesis examines various linear regression methods and their use to establish regression relationships between the compressive strength of concrete determined by the indirect method and by the crushing of the specimens in the press. It deals mainly with the uncertainty of values measured by the indirect method, which is neglected by the usually used ordinary least squares regression method. It also deals with the weighted least squares method, suitable for so-called heteroskedastic data. It compares different regression methods on several sets of previously measured data. The final part of the work examines the effect of removing too influential points identified by Cook's distance, which may skew the regression results.

The aim of this work is to get the best detailed knowledge about hardness distribution in first 60 mm below the surface of backing roll. To this end, a method for obtaining multi-dimensional polynomial regression was developed and then a computer program for its processing was written.Way of finding suitable regression surfaces and their subsequent interpretation, is a pivotal part of this work.

Traumatic brain injury is highly prevalent in the United States yet there is little understanding of how the brain responds during injurious loading. A confounding problem is that because testing conditions vary between assessment methods, brain biomechanics cannot be fully understood. Data mining techniques were applied to discover how changes in testing conditions affect the mechanical response of the brain. Data were gathered from literature sources and self-organizing maps were used to conduct a sensitivity analysis to rank considered parameters by importance. Fuzzy C-means clustering was applied to find any data patterns. The rankings and clustering for each data set varied, indicating that the strain rate and type of deformation influence the role of these parameters. Multivariate linear regression was applied to develop a model which can predict the mechanical response from different experimental conditions. Prediction of response depended primarily on strain rate, frequency, brain matter composition, and anatomical region.

Denna uppsats försöker identifiera faktorer som förklarar varför cykelpendlingen till och från Stockholms innerstad har ökat mellan 1980 och 2020. Vi bildar och väljer ut tre linjära regressionsmodellerna vilka vi anser vara de bästa modellerna utifrån ett flertal urvalskriterier. Resultaten visar att alla inkluderade variabler är signifikanta i respektive modell. Våra resultat, vilka är i linje med tidigare forskning, visar vidare följande: (a) antalet cykelpendlingsresor har ett positivt samband med cykelinfrastrukturkostnader, vilket tyder på att bättre cykelinfrastruktur gör att fler väljer cykeln till jobbet; (b) antalet cykelpendlingsresor har ett positivt samband med befolkningsstorleken; (c) antalet cykelpendlingsresor har ett positivt samband med trängselskatten, vilket tyder på att en överföring sker där en del bilister byter till cykelpendling p.g.a. trängselskatt; (d) antalet cykelpendlingsresor har ett negativt samband med BNP per capita. Sammanfattningsvis indikerar resultaten att ett flertal faktorer har påverkat antalet cykelpendlingsresor till och från Stockholms innerstad mellan 1980 och 2020. / This thesis attempts to identify factors that explain why bicycle commuting in and out of Stockholm City Centre has increased between the years 1980 and 2020. We create and select three linear regression models, which we consider to be the best models based on several selection criteria. Our results show that the included variables in each respective model are significant. Our results, which are in line with previous findings in the literature, moreover, show the following: (a) the number of bicycle commuting trips is positively associated with bicycle infrastructure costs, which indicates that better bicycle infrastructure leads to more people bicycling to work; (b) the number of bicycle commuting trips is positively associated with population size; (c) the number of bicycle commuting trips is positively associated with the congestion tax, which indicates that a modal shift takes place where some motorists switch to bicycle commuting because of the congestion tax; (d) the number of bicycle commuting trips is negatively associated with GDP per capita. In summary, the results indicate that several factors have affected the number of commuting trips by bicycle in and out of Stockholm City Centre between 1980 and 2020.

In this thesis, we set out to model the market risk exposure for 251 stocks in the S&amp;P 500 index, during a ten-year period between 2011-04-30 and 2021-03-31. The study brings to light a model not often mentioned in the scientific literature focused on market risk estimation, the linear mixed model. The linear mixed model makes it possible to model a time-varying market risk, as well as adding structure to the idiosyncratic risk, which is often assumed to be a stationary process. The results show that the mixed model is able to produce more accurate estimates for the market risk, compared to the baseline, which is here defined as a CAPM model. The success of the mixed model, which we in the study will refer to as the ADAPT model (adaptive APT), most certainly lies in its ability to create a hierarchical regression model. This makes it possible to not just view the set of observations as a single population, but let us group the observations into different clusters and in such a way makes it possible to construct a time-varying exposure. In the last part of the thesis, we highlight possible improvements for future works, which could make the estimation even more accurate and also more efficient.

The capacity value (CV) of a power generation unit indicates the extent to which it contributes to the generation system adequacy of a region’s bulk power system. Given the capricious nature of the wind resource, determining wind generation’s CV is nontrivial, but can be understood simply as how well its power output temporally correlates with a region’s electricity load during times of system need. Both wind generation and load are governed by weather phenomena that exhibit variability across all timescales, including low frequency weather cycles that span decades. Thus, a data-driven determination of wind’s CV should involve the use of long-term (i.e., multiple decades) coincident load and wind data. In addition to the challenge of finding high-quality, long-term wind data, existing load data more than several years old is of limited utility due to shifting end usage patterns that alter a region’s electricity load profile. Due to a lack of long-term data, current industry practice does not adequately account for the effects of weather variability in CV calculations. To that end, the objective of this thesis is to develop a model to “hindcast” what the historic regional load in New England would have been if governed by the conjoined influence of historic weather and a more current load profile. Modeling focuses exclusively on summer weekdays since this period is typically the most influential on CV. The summer weekday model is developed using multiple linear regression (MLR), and features a separate hour-based model for eight sub-regions within New England. A total of eighty-four candidate weather predictors are made available to the model, including lagged temperature, humidity, and solar insolation variables. A reanalysis weather dataset produced by the National Aeronautics and Space Administration (NASA) – the Modern Era Retrospective-Analysis for Research and Applications (MERRA) dataset – is used since it offers data homogeneity throughout New England over multiple decades, and includes atmospheric fields that may be used for long-term wind resource characterization. Weather regressors are selected using both stepwise regression and a genetic algorithm(GA) based method, and the resulting models and their performance are compared. To avoid a tendency for overfitting, the GA-based method employs triple cross-validation as a fitness function. Results indicate a regional mean absolute percent error (MAPE) of less than 3% over all hours of the summer weekday period, suggesting that the modeling approach developed as part of this research has merit and that further development of the hindcasting model is warranted.

Machine learning is an area of computer science that only grows as time goes on, and there are applications in areas such as finance, biology, and computer vision. Some common applications are stock price prediction, data analysis of DNA expressions, and optical character recognition. This thesis uses machine learning techniques to predict prices for different stocks and indices on the Swedish stock market. These techniques are then compared to see which performs best and why. To accomplish this, we used some of the most popular models with sets of historical stock and index data. Our best-performing models are linear regression and neural networks, this is because they are the best at handling the big spikes in price action that occur in certain cases. However, all models are affected by overfitting, indicating that feature selection and hyperparameter optimization could be improved.