In large observational studies, data are often highly multivariate with many discrete and continuous variables measured on each observational unit. One often derives subpopulations to facilitate analysis. Traditional approaches suggest modeling such subpopulations with a compilation of interaction effects. However, when many interaction effects define each subpopulation, it becomes easier to model membership in a subpopulation rather than numerous interactions. In many cases, subjects are not complete members of a subpopulation but rather partial members of multiple subpopulations. Grade of Membership scores preserve the integrity of this partial membership. By generalizing an analytic chemistry concept related to chromatography-mass spectrometry, we obtain a method that can identify latent subpopulations and corresponding Grade of Membership scores for each observational unit.

Steel is the worlds most used metal and 2016 the produced amount of steel was measured to a total1630 MT where the Blast Furnace (BF) - Basic Oxygen Steelmaking (BOS) process, which is the mostcommon steel process, stands for 75% of the world’s current crude steel production.The BOS process works by blowing oxygen at high velocity through a hot metal bath consisting of hotmetal from the blast furnace mixed with fluxing agents as slag formers and up to 30 weights% scrap.Due to the high focus on environmental aspects throughout the world, the importance of utilizing andrecycling scrap and by-products such as slag and dust has increased. But when byproducts are recycled,the risk of impurities in the refined hot metal increases. To be able to control the impurities and reducethe risk of getting off-specification end product, it is important to understand when and where theimpurities end up.Zinc and chromium are two of these impurities that may cause issues to the process. To prevent thisfrom happening BlueScope Steel, a steel plant based in Wollongong, Australia, with a productioncapacity of 3 million tons of crude steel, wants to find out where these elements come from and howthey partition. For zinc, it is observed that most of it will end up in the offgas, while for chromium it isnoted that the majority exits with the steel and slag. It is assumed for both elements that the missingamount enters with the shredded steel scrap (shred) to meet the required mass balance using industrialdata. Results showed that the shred needs to contain between 0.312% and 0.515% zinc in order tosatisfy the zinc balance for year 2014 to 2019. For chromium, the range is from 0.315% to 0.371%.Data analyses carried out using SIMCA UMETRICS, which is a multivariate program used to performPrincipal Component Analysis (PCA) and Partial Least Squares (PLS) regression, and normal regressionsdone in Excel showed that zinc in the crude steel cannot be used to calculate or predict the partitioningfactors for zinc between slag, melt and off gases, due to a R2 (Goodness of fit) of 0.144, where the highera R2 value is the better the fit to the model. Slag showed better results with an R2 of 0.473 This regressionwas only done for heats with Zn contents above the detection limit, so it will most likely only beapplicable to heats where relatively high amounts of zinc are expected to end up in the slag. The amountof zinc that ends up in the offgas was above 95% for all the heats modelled.For chromium, results for predicting the partitioning between slag and melt showed greater promise,an R2 of 0.663 was obtained for regression of chromium in crude steel, and an R2 of 0.566 was obtainedfor regression of chromia in the slag. Both values are considered acceptable when analyzing plant data.When comparing the two, using the regression from chromium in the crude steel gives a slightly moreaccurate result, with a partitioning difference of only +/- 0.03 and a 97-99% accuracy, compared to +/-0.05 for chromia in the slag with a 95-98% accuracy.The model is used to compare how different amounts of shred will impact the partitioning of the twoelements analyzed. It also looks at how newer generations of filter cake, containing higher amounts ofzinc will impact the partitioning of zinc between the phases, especially the concentration of zinc thatreports to the dust.

Saber-toothed remains have been found worldwide throughout the Cenozoic, until the end of the Pleistocene. One site from Alachua County, Florida preserves a diverse Miocene fauna, including the machairodontine Nimravides galiani (Felidae) and the saber-toothed Barbourofelis loveorum (Barbourofelidae). Both taxa roamed what would become the Love Bone Bed site during the Late Miocene (Late Clarendonian NALMA), ~ 9.5 Mya. Previous descriptions focused on crania; yet the large sample of postcrania remained undescribed. Hence, this project includes a detailed postcranial description of both taxa. Results show that N. galiani resembles extant felids, whereas B. loveorum resembles Smilodon fatalis, as well as ursids. Additionally, locomotion and hunting behavior (prey capture) was examined quantitatively to assess ecologic overlap (niche partitioning vs direct competition). N. galiani probably displayed terrestrial locomotion in open habitats, whereas B. loveorum likely inhabited the deciduous forest as an ambulatory opportunistic/ambush predator. Results support niche partitioning, rather than direct competition.

One of the main objectives of evolutionary biology is to understand the reasons behind the maintenance of individual differences in a multitude of traits that influence fitness such as locomotor performance and behaviour. Because locomotor performance sets an "envelope" within which behaviour is expressed, it is likely that a multitude of co-adaptations exists between these two suites of traits. In recent years, a growing number of studies have identified associations of different strength and directions between performance and behaviour. Two main hypotheses have received support, on one hand locomotor performance could be "co-specialized" with behaviour in a manner that behaviour reduces predation risk, such that shyer, less active, less explorative animals should be the best sprinters and the most endurant. On the other hand, locomotor performance could "compensate" for behaviours that lead to increased predation risk, in a way that bolder, more active and explorative animals should be able to sprint faster and for longer. In my thesis I provide a review of published studies that successfully identify associations between locomotor performance and behaviour and classify each association as supporting the co-specialization or compensation hypothesis respectively. I further elaborate on the importance of using repeated measurements and (co)variance partitioning when studying correlations between labile traits. I also discuss one of the main challenges that comes with studying locomotor performance, namely the importance of the variation in motivation, both methodologically, by using different performance tests, but also physiologically, by using blood corticosterone measurements as indicators of such variation.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.