Doctor of Philosophy / Family Studies and Human Services / Jared R. Anderson / Cyclical romantic relationships—those characterized by breaking up and getting back together or having on/off periods—are a frequent phenomenon in the emerging adult population. These dating relationships maintain some distinctions from other more stable relationships, including the ways that partners strive to sustain relationship health. The purpose of this phenomenological qualitative inquiry was to increase in-depth understanding of how emerging adult dating partners’ relationship effort affects relationship transitions within cyclical dating relationships. Ten heterosexual emerging adult couples (10 men, 10 women) currently in cyclical dating relationships were interviewed about their experiences with relationship effort and maintenance. Participant interviews were analyzed according to the Interpretative Phenomenological Analysis (IPA) method. Specific themes emerged from the data, focusing on how perceived individual effort in the relationship, perceived partner effort in the relationship, and specific maintenance behaviors couples used to sustain relational health affected couple decisions about relationship transitioning. Implications regarding relationship education and clinical intervention among cyclical emerging adult couples are discussed. Future research could focus on continued expansion of understanding when in relationship history cyclical patterns begin, and how partners navigate transitions when both perceive reduced relationship effort.

Zugl.: Würzburg, Universiẗat, Diss., 2008.

This dissertation studies the effects of credit availability on firm-level outcomes using a new matched employer-employee panel of Chilean firms that also includes firm-bank lending and tax data. In Chapter I, using a natural experiment and a differences in differences approach, I show that firms that experienced a positive credit supply shock during the 2008-09 recession in Chile, exhibit higher labor productivity four years after the shock, even after aggregate demand and credit supply have fully recovered. Chapter II presents evidence consistent with the hypothesis that at least part of the productivity improvement is due to an increased ability of firms with access to credit to adjust labor during the recession. In particular, I find that these firms exhibit larger worker flows and use credit to adjust employment by churning more workers. Chapter III studies a government partial credit guarantee scheme for bank loans to small and medium sized enterprises using a regression discontinuity design around the threshold for eligibility. I show that the program has a large positive causal effect on firms' total borrowing, and the effect is persistent. Moreover, firms that obtain bank loans through this scheme can borrow more from loans not insured by the guarantee, which means that the program has a positive effect on the firms's total borrowing capacity. Finally, the program also helps in the formation of new bank-firm lending relationships.

Philosophiae Doctor - PhD / This study examines the problems of implementing water allocation policy in the context of the local state bureaucracy as well as the specific experiences of local black emerging farmers in the Breede Gouritz Water Management Area. This study used qualitative research methods and is based on many hours of interviews and observing bureaucrats and stakeholders at the receiving end of the bureaucratic business process of water allocation. It is not only concerned with the physical and technical aspects of access but explores how the different role players interact, navigate, shape, frame and manage challenges to gain access to and control water for productive use. The actual experiences and understandings of the stakeholders in their own contexts when engaging with the access to water are crucial to gain a comprehensive understanding and insight into the influence of bureaucracy and power relations. This thesis therefore maps the confusions and incapacities and shows that even though the South African laws are based on the best international frameworks, they fail, as they do not sufficiently address the unique environment and landscape. Existing scholarship has not adequately researched local bureaucratic power. At the coalface of implementation, bureaucrats make up their own rules to cope with rapid policy churning. Combined with existing power relations, policy implementation and policy direction is steered towards different and unintended trajectories, making transformation a challenge to achieve. Consequently, my main finding is that there have been constant and rapid legislative and policy changes but they have simply added to the confusion and instability.

The rapid growth of technological infrastructure has changed the way companies do business. Subscription based services are one of the outcomes of the ongoing digitalization, and with more and more products and services to choose from, customer churning has become a major problem and a threat to all firms. We propose a machine learning based churn prediction model for a subscription based service provider, within the domain of financial administration in the business-to-business (B2B) context. The aim of our study is to contribute knowledge within the field of churn prediction. For the proposed model, we compare two ensemble learners, XGBoost and Random Forest, with a single base learner, Naïve Bayes. The study follows the guidelines of the design science methodology, where we used the machine learning process to iteratively build and evaluate the generated model, using the metrics, accuracy, precision, recall, and F1- score. The data has been collected from a subscription-based service provider, within the financial administration sector. Since the used dataset is imbalanced with a majority of non- churners, we evaluated three different sampling methods, that is, SMOTE, SMOTEENN and RandomUnderSampler, in order to balance the dataset. From the results of our study, we conclude that machine learning is a useful approach for prediction of customer churning. In addition, our results show that ensemble learners perform better than single base learners and that a balanced training dataset is expected to improve the performance of the classifiers.

No description available.

Lately the discussion on how big impact vehicles has on the environment has grown bigger and bigger. The vehicle manufacturers are building cars, trucks, and wheel loaders etc. that are getting more and more fuel efficient for every year. But where do the engineers at the companies make the improvements? The drive train is one area of the vehicle where many different sources of losses are found. The gearbox is one example, where the losses are both dependent and independent on the torque transferred by the gear pairs. The gearbox consists of several components that all of them contribute to the total system power loss. It is therefore of great interest to understand the magnitude of the losses in the gears, for different operating conditions, and also to be able to determine how they change with, for instance, rotational speed, oil viscosity. The goal of the master thesis was to make a study about the oil churning losses in a 6 speed automatic gearbox. This model makes use of calculations which were based on scientific experiments. This makes the loss calculations more advanced and accurate than the generalized formulas in most commercial software’s. This model takes into account only the oil churning and oil pocketing losses and shall not hold account for the windage losses or the other losses such as losses occurring due to bearings, seals, pumps etc. Due to time constraint a user friendly programme could not be written but the generated programme can be modified to make it easier for the user to input the data and produce results. All gears are modelled in simple modules built using MATLAB m-codes. Assumptions are made in the code when it came to choosing the oil viscosity. / På senare år har diskussionen om fordonens miljöpåverkan blivit allt mer intensiv. Fordonstillverkarna bygger bilar, lastbilar, hjullastare, etc., vars bränsleeffektivitet årligen når nya höjder. Var finns utrymme för förbättringar? Drivlinan är en del av fordonet där många olika förlustkällor påträffas; t.ex. växellådan, där förlusterna både är beroende och oberoende av det vridmoment som överförs av kugghjulsparen. Växellådan består av ett flertal komponenter som bidrar till systemets totala effektförlust. Det är därför av stort intresse att förstå omfattningen av förlusterna i kugghjulen, för olika driftsförhållanden, och att kunna bestämma hur förlusterna är relaterade till t.ex. rotationshastighet och oljeviskositet. Målet med examensarbetet var att göra en undersökning om oljeskvalpförluster (”oil churning losses”) i en 6-växlad automatlåda. Modellen använder beräkningar baserade på vetenskapliga experiment. Detta gör förlustberäkningarna mer avancerade och exakta än de allmänna formler som de flesta kommersiella programvaror nyttjar. Modellen tar endast hänsyn till oljeskvalpförluster och oljeficksförluster (”oil pocketing losses”) och tar icke hänsyn till vindförluster eller andra förluster såsom lagerförluster, friktion i tätningar, pumpförluster, etc. På grund av projektets tidsramar har förfiningen av programmets användargränssnitt prioriterats lågt; vid behov kan detta åtgärdas för att förenkla input och output av data. Alla kugghjul modelleras i enkla moduler som byggts som MATLAB-skript. Antaganden görs i koden när det gäller oljans viskositet .

This report is about examining the fields of machine learning and digital marketing, using machine learning as a tool to predict churn in a new domain of companies that do not track their customers extensively, i.e where behaviour data is discontinuous.  To predict churn relatively simple out of the box models, such as support vector machines and random forests, are used to achieve an acceptable outcome. To be on par with the models used for churn prediction in subscription based services, this report concludes that more research has to be done using more effective evaluation metrics. Finally it is presented how these discoveries can be commercialized and the business related benefits of using churn prediction for the employer Sellpy. / Denna rapport handlar om att utforska fälten maskininlärning och digital marknadsföring, genom att använda maskininlärning som ett redskap för att förutspå churn i en typ av företag med diskontinuerlig beteendedata. För att förutspå churn finns relativt simpla "out of the box"-modeller, som support vector machines och random forests, som används för att nå acceptabla resultat. För att nå liknande resultat som i arbeten där churn utförs på kontinuerlig beteendedata konstaterar denna rapport att framtida arbeten forska på vilka utvärderingsmetriker som är mest lämpade. I rapporten presenteras också hur dessa upptäckter kan kommersialiseras och hur företaget Sellpy kan tjäna på att förutspå churn.

No description available.

No description available.