M.S.

Yes / This note highlights the value of using UQR for addressing the limitations inherent within previous methods involving conditional parameter distributions for spending analysis (QR and OLS). Using unique data and robust analysis using improved methods, our paper clearly demonstrates the over-importance attached to length of stay and the inadequate attention given to business travelers in previous research. There are clear benefits from UQR’s methodological robustness for assessing the multitude of variables related to tourist expenditures, particularly given UQR’s ability to inform across the spending distribution. Given tourism’s importance for the UK it is critical for expensive promotional activities to be targeted efficiently for ensuring effective policy making.

Yes / We employ unconditional quantile regression with region of origin fixed effects, whereby we find that attending live football matches significantly increases expenditures by inbound tourist in the UK, and surprisingly we find that such effects are strongest for those who overall spend the least. Higher spending individuals spend significantly more than those who do not attend football matches, even when such individuals are otherwise similar. We analyse the impact of football attendance across the tourism expenditure distribution which is a relatively neglected aspect within previous research.

Yes / The inbound tourist expenditure generating role of football (soccer), particularly the English Premier League 15 (EPL) is evaluated. An enhanced economic and management understanding of the role of regular sporting fixtures emerges, as well as quantification of their impact. Expenditure on football tickets is isolated to identify local economic spillovers outside the stadium walls. Using the UK International Passenger Survey, unconditional quantile regressions (UQR) is used to evaluate the distributional impact of football attendance on tourist expenditures. Both total expenditure and a new measure which adjusts expenditures for football ticket prices are considered. UQR is a novel technique which is as yet underexploited within sport economics and confers important methodological advantages over both OLS and quantile regressions. Significant cross quantile variation is found. High spending football fans spend more, even after ticket prices are excluded. Surprisingly, spending effects owing to attendance are strongest for those who overall spend the least, confirming the role of sport as a generator of tourist expenditure unlike most others. Though the attendance effect is smaller for higher aggregate spenders, there is nevertheless a significant impact across the distribution. Distributional expenditure impacts highlight clear differentials between attendance by high and low spenders. Similar analysis is applicable to other global brands such as the National Football League (NFL) in the United States (American football) and the Indian Premier (cricket) League. The EPL’s global popularity can be leveraged for achieving enhanced tourist expenditure.

Yes / penalized is a exible, extensible, and e cient MATLAB toolbox for penalized maximum likelihood. penalized allows you to t a generalized linear model (gaussian, logistic, poisson, or multinomial) using any of ten provided penalties, or none. The toolbox can be extended by creating new maximum likelihood models or new penalties. The toolbox also includes routines for cross-validation and plotting.

This paper applies and extends Machado and Mata's parametric quantile decomposition method and a similar nonparametric technique to explore changes in welfare in Zimbabwe between 1995 and 2001. These methods allow us to construct a counterfactual distribution in order to decompose the shift into the part due to changes in endowments and that due to changes in returns. We examine two subsets of a nationally representative dataset and find that endowments had a positive effect but that returns account for more of the difference. In communal farming areas, the effect of returns was positive while, in urban Harare, it was negative. / Master of Science

The purpose of this thesis is to predict the lifespan of a battery using a predictive model, utilizing data from early cycles. The goal is to minimize both time and costs for the company by reducing the number of cycles needed for testing. Currently, the company tests a diverse set of batteries, which is both time and resource-consuming. To investigate which data-driven predictive model should be used by the company to predict battery capacity at XX cycles, a thorough literature study has been conducted. In summary, a variety of variables from specific cycles have been calculated based on inspiration from Fei et al. (2021), Severson et al. (2019), Enholm et al. (2022) and an internal project from the company. Following this, two different predictive models, Gaussian Process Regression and Ordinary Least Squared Regression, are applied and compared.  Based on the obtained results, Gaussian Process Regression had a slight better results but a significantly higher complexity compared to Ordinary Least Squared Regression. Therefore, the data-driven model that should be implemented at the company is an Ordinary Least Squared Regression with variables related to different phases during a cycle. This result is primarily based on the varying degrees of complexity of the models. / Syftet med detta examensarbete är att med hjälp av en datadriven prediktionsmodell kunna prediktera livslängden på ett batteri genom att använda data från tidiga cykler. Målet är att minimera både tid och kostnader för företaget genom att minska antalet cykler som behövs för testning. I dagsläget testar företaget en mängd batterier vilket både är tids- samt resurskrävande. För att undersöka vilken datadriven prediktionsmodell som bör användas av företaget för att prediktera batteriekapacitet vid XX cykler har en gedigen litteraturstudie utförts. Sammanfattningsvis har en mängd variabler av de mätningar som finns från specifika cykler beräknats utifrån inspiration från Fei med flera (2021), Severson med flera (2019), Enholm med flera (2022) samt ett internt projekt från företaget. Efter detta applicerades och jämfördes två olika prediktionsmodeller: Gaussian Process Regression och Ordinary Least Squared Regression.  Baserat på de erhållna resultaten hade Gaussian Process Regression något bättre resultat men en betydligt högre komplexitet jämfört med Ordinary Least Squared Regression. Därför är den datadrivna modell som bör implementeras på företaget en Ordinary Least Squared Regression med variabler relaterade till olika faser under en cykel. Detta resultat grundar sig framför allt i olika grad av komplexitet hos modellerna.

Master of Science / Department of Statistics / Paul I. Nelson / The principle of least squares applied to regression models estimates parameters by minimizing the mean of squared residuals. Least squares estimators are optimal under normality but can perform poorly in the presence of outliers. This well known lack of robustness motivated the development of alternatives, such as least median of squares estimators obtained by minimizing the median of squared residuals. This report uses simulation to examine and compare the robustness of least median of squares estimators and least squares estimators of the slope of a regression line through the origin in terms of bias and mean squared error in a variety of conditions containing outliers created by using mixtures of normal and heavy tailed distributions. It is found that least median of squares estimation is almost as good as least squares estimation under normality and can be much better in the presence of outliers.

Das Verkehrsmittelwahlverhalten von Menschen ist komplex. So spielen soziodemografische, sozioökonomische sowie raum- und siedlungsstrukturelle Merkmale eine Rolle. In dieser Arbeit wird dieses Verhalten untersucht. Dabei wird eine homogene Grundgesamtheit geschaffen, welche alle Personen beinhaltet, die sowohl über eine Dauerkarte des öffentlichen Personenverkehrs als auch einen Personenkraftwagen verfügen. Anhand derer soll eine deskriptive Analyse und eine multinomiale logistische Regression Aufschluss geben, ob es Unterschiede zwischen den jeweiligen Nutzergruppen gibt. So lässt sich die Gruppe der ÖV-Nutzer durch folgende Charakteristiken beschreiben: der Großteil sind Frauen, sowie Personen, die eine hohe schulische und berufliche Bildung besitzen. Des Weiteren werden eher weniger Wege mit dem ÖV als mit dem PKW zurückgelegt. Erwerbstätige hingegen entscheiden sich eher für den PKW. / Human behavior towards the choice of transportation varies in very complex ways such as sociodemographics, socioeconomics as well as settlement structures. For this paper a homogenous population is created from season ticket holders for public transportation and car owners. Based on this population a descriptive analysis followed by a multinomial logistic regression is supposed to generate the differences between the user groups. The group of users of the public transportation system can be characterized as followed: the majority of users are women as well as highly educated people. Within this specific group distances are more likely to be covered by public transportation rather than by car. However the working population prefers to go by passenger car.

This dissertation consists of two chapters. The first chapter establishes an algorithm for calculating capital requirements. The calculation of capital requirements for financial institutions usually entails a reevaluation of the company's assets and liabilities at some future point in time for a (large) number of stochastic forecasts of economic and firm-specific variables. The complexity of this nested valuation problem leads many companies to struggle with the implementation. The current chapter proposes and analyzes a novel approach to this computational problem based on least-squares regression and Monte Carlo simulations. Our approach is motivated by a well-known method for pricing non-European derivatives. We study convergence of the algorithm and analyze the resulting estimate for practically important risk measures. Moreover, we address the problem of how to choose the regressors, and show that an optimal choice is given by the left singular functions of the corresponding valuation operator. Our numerical examples demonstrate that the algorithm can produce accurate results at relatively low computational costs, particularly when relying on the optimal basis functions. The second chapter discusses another application of regression-based methods, in the context of pricing variable annuities. Advanced life insurance products with exercise-dependent financial guarantees present challenging problems in view of pricing and risk management. In particular, due to the complexity of the guarantees and since practical valuation frameworks include a variety of stochastic risk factors, conventional methods that are based on the discretization of the underlying (Markov) state space may not be feasible. As a practical alternative, this chapter explores the applicability of Least-Squares Monte Carlo (LSM) methods familiar from American option pricing in this context. Unlike previous literature we consider optionality beyond surrendering the contract, where we focus on popular withdrawal benefits - so-called GMWBs - within Variable Annuities. We introduce different LSM variants, particularly the regression-now and regression-later approaches, and explore their viability and potential pitfalls. We commence our numerical analysis in a basic Black-Scholes framework, where we compare the LSM results to those from a discretization approach. We then extend the model to include various relevant risk factors and compare the results to those from the basic framework.