Search

People constantly attribute positive or negative qualities to objects in their environment. What the organism considers to be positive or negative depends on the context in which it currently finds itself. While objects may be perceived as positive in one context, they may appear negative in another context. The social environment in which we operate is generally more positive than negative. In many cases, this positive environment results in negative information having advantages over positive information (i.e., a so-called ‘negativity bias’). For example, negative information tends to be better remembered. Interestingly, there’s hardly any evidence for this otherwise well-documented advantage of negative information in experiments on Evaluative Conditioning (EC). EC is defined as the change in liking of a conditioned stimulus (CS) due to its pairing with a positive or negative unconditioned stimulus (US). One reason for the absence of a negativity bias in EC experiments might be their symmetric structure, which does not reflect the information ecology of the real world. Other similar structural differences could also explain why typical EC experiments consistently elicit EC effects in the laboratory, while EC rarely occurs in the real world. In this thesis, I investigate when EC effects occur, focusing on the interplay between environment and organism. Additionally, I examine whether the difference in the structure of the real world and EC experiments can explain the absence of a negativity bias in EC experiments. The findings show that, in addition to the environment, the organism plays a crucial role in perceiving CS and US pairings. Furthermore, the thesis provides evidence that a cognitive-ecological approach can contribute to explaining the absence of a negativity bias in EC.

This dissertation mainly deals with the question of how social identities affects individual behavior in the domains of social status-seeking behavior, belief formation processes and migration preferences.

Die Universitäts- und Stadtbibliothek Köln (USB) entwickelt in enger Zusammenarbeit mit dem Cologne Center for eHumanities (CCeH) und dem Philosophischen Seminar der Universität zu Köln, den von der Deutschen Forschungsgemeinschaft (DFG) geförderten Fachinformationsdienst(FID) Philosophie. Dessen Ziel ist es, den Spitzenbedarf der deutschsprachigen Philosophie in seiner gesamten systematischen Breite abzudecken. Dabei versteht sich der FID Philosophie als Informationsdienstleister, der die Fachcommunity aktiv in seine Planungen einbezieht. Mit philportal.de hat der FID Philosophie ein umfangreiches Serviceangebot konzipiert, das diesen Anforderungen perspektivisch entsprechen soll und stetig weiterentwickelt wird. Das Portal bietet neben einer Rechercheplattform für philosophische Fachliteratur auch einen umfassenden Zugriff auf elektronische Volltexte an. Zudem unterstützt der FID Wissenschaftler*innen und Verlage bei der Einrichtung oder der Migration von Open Access Journals. Auch im Bereich des Forschungsdatenmanagements ist das Projekt aktiv: Neben einem umfangreichen Angebot an Workshops arbeitet der FID an der Sichtbarmachung von Forschungsdaten in bibliothekarischen Nachweissystemen und vermittelt Forschende an Expert*innen innerhalb der Nationalen Forschungsdateninfrastruktur (NFDI). Aufgrund der Disziplin inhärenten fachübergreifenden Tendenzen zeichnet sich die Community durch ihre Heterogenität aus, sodass die jeweiligen Bedarfe der Forschenden hoch divers sind. Um die konzeptionellen Maßnahmen des FID Philosophie sowie die Entwicklung des PhilPortals weiterhin zielgerichtet an den aktuellen Bedarfen der philosophischen Fachcommunity anpassen und die Informationsversorgung nachhaltig aufrechterhalten zu können, ist eine regelmäßige Evaluation der sich dynamisch entwickelnden Anforderungen an das Projekt zwingend erforderlich. Die vorausgegangene Community-Befragung und die im Anschluss erfolgte Evaluation der Ergebnisse sollen sicherstellen, dass die Bed

The thesis contains three self-contained essays on statistical methods for the analysis of financial risk. First, we compare alternative estimation approaches for factor augmented panel data models that are also relevant in the field of financial risk. Our focus lies on panel data sets where the number of panel groups (N) is large relative to the number of time periods (T). The Principal Component (PC) and Common Correlated Effects (CCE) estimators were originally developed for panel data with large N and T, whereas two considered GMM approaches assume that T is fixed in the asymptotic analysis. Our comparison of existing methods addresses three different issues. First, we analyze the possibility of an inappropriate normalization of the factor space (the so-called normalization failure). In particular we propose a variant of the CCE estimator that avoids the normalization failure by adapting a weighting scheme inspired by the Mundlak approach. Second, we analyze the effects of estimating versus fixing the number of factors in advance. Third, we demonstrate how the design of the Monte Carlo simulations favors some estimators, which explains the conflicting findings from existing Monte Carlo experiments. Second, we consider empirical challenges that are relevant for portfolio selection. Portfolio selection according to Markowitz requires reliable estimates of the expected returns and the covariance matrix of returns. Estimating these moments via their sample analogs (the so-called plug-in method) yields extreme portfolio weights that excessively fluctuate over time and typically perform poorly out-of-sample. Obtaining reliable estimates is particularly problematic in case the number of investable assets is of similar magnitude as the available amount of time series data. A common procedure is to ignore information on the first moment, resulting in the estimation of the global minimum variance (GMV) portfolio, which generally improves the out-of-sample performance. In

Das maligne Melanom befindet sich unter den 5 häufigsten Krebserkrankungen Deutschlands und es ist einer der Tumorerkrankungen mit der höchsten Neigung zur zerebralen Metastasierung. Neben der Strahlentherapie, welche immer häufiger als stereotaktische/radiochirurgische Bestrahlung erfolgt, ist die medikamentöse Behandlung ein weiterer wichtiger Bestandteil der Therapie. Seit der Einführung bestimmter zielgerichteter, medikamentöser Therapien im Jahr 2010 wird standardmäßig der Mutationsstatus des malignen Melanoms untersucht, um wirkungsvollere Behandlungen anbieten zu können. Die vorliegende retrospektive, statistische Arbeit untersucht den Einfluss des BRAF-Mutationsstatus auf die lokale Kontrolle nach stereotaktischer Bestrahlung zerebraler Metastasen bei malignem Melanom. Bei der Datenaquise in der Klinik und Poliklinik für Radioonkologie, Cyberknife- und Strahlentherapie der Uniklinik Köln, im Zeitraum vom 01.01.2010 bis 31.12.2016, wurden bei 93 Patienten 86 BRAF-Wildtyp und 150 BRAF-mutierte Metastasen identifiziert. Um die Kohorten zu charakterisieren, wurden sie bezüglich Geschlechts, Alter, Mutationsstatus, Gesundheitszustand, medikamentöser und zusätzlicher Ganzhirnbestrahlung verglichen. Hier zeigte sich, dass das mediane Erkrankungsalter von BRAF mutierten Metastasen mit 51 Jahren signifikant geringer ist als das bei BRAF Wildtyp Metastasen mit 74 Jahren. Die Daten konnten keinen signifikanten Einfluss des Mutationsstatus auf die lokale Kontrolle nachweisen (p=0,579). Die lokale Kontrollrate liegt bei BRAF mutierten Metastasen bei knapp über 40% und bei BRAF-Wildtyp Metastasen bei ca. 50%. Dahingegen ist das Gesamtüberleben für BRAF-mutierte Metastasen mit median 13 Monaten signifikant länger als jenes von BRAF-Wildyp Metastasen mit 8 Monaten (p<0,001). Eine Erklärung hierfür bietet die medikamentöse Behandlung in Form der zielgerichteten Therapie. Die Ergebnisse reihen sich in die vorhandenen Publikationen zu diesem Thema ein. Eine mögliche bev

In dieser Studie wurde ein neuer Cancellation Test (Find-the-Symbol-Test; FST) mit seinen fünf Parallelversionen (A-E) anhand einer gesunden Studienpopulation (N=192) normvalidiert und anschließend an Hirntumorpatient*innen (N=20) getestet. Zusätzlich führten die Testpersonen drei andere etablierte Neglekttests durch (Bells Test, Apples Test, Line Bisection Test). Die insgesamt vier Neglekttests wurden zunächst hinsichtlich des Einflusses verschiedener demographischer Variablen untersucht (Alter, Geschlecht, Bildung). Beim FST wurde zusätzlich jede der fünf Parallelversionen (A-E) getrennt analysiert. Hierbei wurden drei Parameter betrachtet: Gesamtergebnis (Anzahl korrekt markierter Symbole), Asymmetriescore (Differenz der rechts- und linksseitig markierten Symbole), Bearbeitungszeit. Beim Gesamtergebnis des FST hatten sowohl das Geschlecht als auch das Alter einen signifikanten Einfluss (p=0,005; p≤0,001) auf das Testergebnis. Die weiblichen Testpersonen sowie die jüngeren Altersgruppen schnitten signifikant besser ab als die männlichen Testpersonen und die älteren Altersgruppen. Beim Asymmetriescore des FST war nur der Einfluss des Alters signifikant (p=0,002). Bei den Parallelversionen A-E gab es Unterschiede hinsichtlich des Einflusses von Geschlecht und Altersgruppe auf die Testergebnisse. Der Bildungsstand und die Reihenfolge, in der die Tests bearbeitet wurden, hatten bei keiner Version einen signifikanten Einfluss auf die Testergebnisse. Bei den etablierten Neglekttests (Bells Test, Apples Test, Line Bisection Test) konnte nur beim Line Bisection Test ein signifikanter Einfluss von Alter und Geschlecht auf die Testergebnisse festgestellt werden (Gesamtergebnis: p=0,024; p=0,018 vs. Asymmetriescore: p=0,010; 0,046). Die männlichen Testpersonen schnitten hier signifikant besser ab als die weiblichen, die älteren Testpersonen erzielten ein schlechteres Ergebnis. Auffallend war, dass die Bearbeitungszeiten aller Tests signifikant vom Alter und dem Bildungsstand

The core objective of this cumulative dissertation is to generate new insights in the occurrence and prediction of unethical firm behavior disclosure. The first two papers investigate predictors and antecedents of (severe) unethical firm behavior disclosure. The third paper addresses frequently occurring methodological issues when applying machine learning approaches within marketing research. Hence, the three papers of this dissertation contribute to two recent topics within the field of marketing: First, marketing research has already focused intensively on the consequences of corporate misconduct and the accompanying media coverage. Meanwhile, the prediction and the process of occurrence of such threatening events have been examined only sporadically so far. Second, companies and researchers are increasingly implementing machine learning as a methodology to solve marketing-specific tasks. In this context, the users of machine learning methods often face methodological challenges, for which this dissertation reviews possible solutions. Specifically, in study 1, machine learning algorithms are used to predict the future occurrence of severe threatening news coverage of corporate misconduct. Study 2 identifies relationships between the specific competitive situation of a company within its industry and unethical firm behavior disclosure. Study 3 addresses machine learning-based issues for marketing researchers and presents possible solutions by reviewing the computer science literature.

Over the past decades, various methods have been used to scan the human genome to identify genetic variations associated with diseases, in particular with common, complex disorders. One of such approaches is the genome-wide association study (GWAS), which compares genetic variation between affected and healthy individuals to find genomic variants in the DNA sequence associated with a trait. GWAS are usually conducted separately for individual traits, and the same single nucleotide polymorphisms (SNP)/loci are associated with different traits in independent studies 7-10. These findings buttress the knowledge that most complex traits are correlated and have shared genetic architecture, therefore, sharing the same heritable risk factors11. Knowledge of the genetic risk factors can directly or indirectly contribute to improvements in risk assessment, drug target development, and ultimately in providing effective therapies to the affected individuals. Pleiotropy is the phenomenon of a hereditary unit affecting more than one trait, and the earliest reported evidence was provided by Mendel when he noted that some set of features were always observed together in a plant. Although this example could have been purely due to linkage and could be regarded as spurious pleiotropy in recent times, it opened up more discussion and research into pleiotropy, which has since been an active area of research12. In this work, I focused on complex epilepsies and the overlap in the genetic factors impacting their phenotypes. Epilepsy is a brain disorder comprising monogenic and common/complex forms characterized by recurrent partial or generalized seizures. However, the extent to which genetic variants contribute to the disorder and how much of the genetic contribution is shared between the different phenotypes is not yet fully understood. This motivated this project, where I benchmarked available pleiotropy detection approaches to select the best performing method in terms of power and fa

This dissertation comprises five essays on advanced digital financial reporting formats, their determinants, and their impact on capital markets. The first chapter leads through the outline of the dissertation, introduces the research context, and provides a framework for the topic of advanced digital reporting formats in research and practice. Chapters two and three analyze the usage and empirical determinants of the voluntary usage of online financial reports (OFR) based on HTML and its impact on the information environment in Europe. In the fourth chapter, the dissertation analyzes the quality of OFR and its corporate governance determinants. Chapter five deals with a literature review on the adoption of XBRL worldwide, its potential impact, and respective influencing factors. The closing chapter comprises a qualitative research study based on semi-structured interviews with experts from large, listed firms as well as auditing and advisory companies in Germany in the context of the mandatory adoption of digital reporting formats and tackles the topics of organizational and processual integration as well as financial reporting and communication considerations.

High-performance virtual reality (VR) technology has opened new possibilities for the examination of the reactivity towards addiction-related cues (cue-reactivity) in addiction. We conducted two studies employing a virtual reality cue-reactivity design for participants suffering from gambling disorder (GD), that combined the assessment of subjective, physiological, and behavioral cue-reactivity. The first study aimed to examine the reliability of temporal discounting measures in VR and standard lab environments in a group of non-gambling control participants. Additionally, we aimed to explore the feasibility of applying sequential sampling models to temporal discounting data obtained in VR. The second study employed the VR design validated in the first study to investigate the subjective, behavioral, and physiological effects of VR gambling environment exposure in a group of regular gamblers (GD group) and matched non-gambling controls. Overall, the results obtained by both studies presented in this dissertation project revealed further evidence for the validity of temporal discounting and the two-step task as possible diagnostic markers of GD. We demonstrated high reliability of the temporal discounting task and reproduced established group differences in decision-making between participants suffering from GD and non-gambling controls in both behavioral tasks. Additionally, we showed that behavioral data obtained by both tasks in VR can be meaningfully interpreted with comprehensive computational modelling, especially with models including RTs such as the drift-diffusion model. In the context of cue-reactivity we found mixed results. While our design was effective in eliciting subjective craving in participants suffering from GD, we observed little evidence for behavioral or sympathetic physiological cue-reactivity.

This thesis addresses the modeling and prediction of portfolio weights in high-dimensional applications to returns on a set of risky financial assets with a particular emphasis on Global Minimum Variance Portfolio (GMVP) allocations. In total, it comprises three self-contained essays on modeling and predicting dynamic portfolio weights for financial asset returns. The first essay is a joint paper with Prof. Dr. Roman Liesenfeld and Jun.-Prof. Dr. Fabian Krüger. The second essay is a joint work with Prof. Dr. Roman Liesenfeld and Prof. Dr. Guilherme Valle Moura. The third essay is a single authored project. In the first essay a novel dynamic approach to forecast the weights of the global minimum variance portfolio (GMVP) for the conditional covariance matrix of asset returns is proposed. Building on a representation in which the GMVP weights obtain as population coefficients of a linear regression of a benchmark return on a vector of return differences, a consistent loss function is derived from which the GMVP weights can be inferred without imposing any distributional assumptions on the returns. In order to capture time variation in the returns' conditional covariance structure, the portfolio weights are modeled through a recursive least squares (RLS) scheme as well as by generalized autoregressive score (GAS) type dynamics. These models are applied to daily stock returns, and it is found that they perform well compared to existing static and dynamic approaches in terms of both the expected loss and unconditional portfolio variance. In the second essay, again the regression representation of the GMVP weights is the starting point, from which a linear state space model with time-varying parameters and stochastic volatility is derived. This specification allows to address both the time variation in the assets' conditional covariance structure and the heteroscedasticity in the market. Using Bayesian regularization techniques, the portfolios are robustified against esti

This thesis consists of two independent parts. In the first part, we investigate the notion of locality in the context of Matrix Product States (MPS), while in the second part, we minimise the statistical error in a calibration protocol for quantum-gate sets. In the first part of this thesis, we consider an MPS in a one-dimensional lattice and investigate its norm squared amplitudes with respect to a local orthonormal basis. We call this norm the classical restriction of the MPS. Concretely, we ask when the classical restriction of an MPS is quasi-locally Gibbsian, i.e., it is exponentially well approximated by a Gibbs distribution associated with a local Hamiltonian. We prove that the classical restriction of an injective MPS is quasi-locally Gibbsian if the matrices associated with the MPS satisfy a ‘purity’ condition, a condition previously established in the theory of random matrix products. Our result connects two notions of locality: locality of correlations in an MPS and locality of interactions in the Hamiltonian generating the corresponding Gibbs distribution. The proof of our result consists of two steps. First, given an MPS defined on a lattice for which the purity condition holds, we demonstrate that the classical Conditional Mutual Information (CMI) of any connected tripartition of the lattice is rapidly decaying in the width of the middle region. Then, this decaying of the classical CMI is shown to imply that the probability distribution associated with the classical restriction of the MPS is quasi-locally Gibbsian. Within this investigation, we present relevant observations around the purity condition. We research how ‘typical’ the purity condition is by constructing a probabilistic model and showing that, in this model, the purity condition is satisfied in general. Furthermore, we show that violating the purity condition enables a generalised notion of error correction, reinforcing the purity condition’s generic nature. Within the second part of this

In this cumulative dissertation, statistical models for regression are discussed in light of high-dimensional, biological data. The dissertation includes three publications: RNA transcription and degradation of Alu retrotransposons depends on sequence features and evolutionary history examines Alu elements, RNA retrotransposons in the human genome. Their RNA metabolism is poorly understood, and the source of Alu transcripts is still unresolved. We have conducted a transcription shutoff experiment and metabolic RNA labelling to shed further light on the life cycle of Alu transcripts.We furthermore present a novel statistical test for detecting expression quantitative trait loci relying on k-mer sequence representation. Endoscopic hemostasis makes the difference: Angiographic treatment in patients with lower gastrointestinal bleeding uses retrospective study data from patients receiving either endoscopic or angiographic treatment for lower gastrointestinal bleeding. While a majority of patients can be treated successfully with the usually preferred endoscopic method, in some cases, angiography is required to achieve hemostasis. Using conditional inference trees, we construct a decision tree model predicting if a patient should receive angiographic treatment. Genetic instability and recurrent MYC amplification in ALK-translocated NSCLC: a central role of TP53 mutations investigates a molecular subtype of lung cancer exhibiting rearrangements of the ALK gene. This cancer type often resists treatments, and no reliable biomarker to identify patients at risk for relapse is known. Analysing biopsy and cell culture data, we find that mutations in the TP53 gene can lead to chromosomal instability and thus the amplification of known cancer genes. This, in turn, grants cancer cells a proliferative advantage compared to the wild-type, providing a new approach for diagnosis and treatment.

In dieser Arbeit werden Stornierungen von Hotelbuchungen mithilfe verschiedener Methoden maschinellen Lernens vorhergesagt. Buchungsstornierungen stellen für die Hotelindustrie ein Problem dar, da sie die Planbarkeit von Einnahmen und Kosten stark erschweren. Unter Verwendung von logistischen Regressionen, Random Forests und künstlichen neuronalen Netzen werden Modelle erstellt, die diese Stornierungen vorhersagen sollen. Diese verschiedenen Modelle werden anschließend miteinander verglichen, um die effektivste Methode zu finden. Vorhersagemodelle könnten bestehende Probleme, die durch Stornierungen entstehen, lösen, in dem Maßnahmen wie Überbuchungen effektiver eingesetzt werden könnten. Bei der Analyse wird ein Datensatz, der von zwei portugiesischen Hotels stammt, mit fast 120.000 realen Beobachtungen und 31 Variablen zu Hilfe genommen., The aim of this thesis is to predict hotel booking cancellations with different machine learning methods. Hotel booking cancellations are a problem for hotels, because they diminish revenue and make it harder to plan future income. Using logistic regressions, Random Forests and artificial neural networks different prediction models are computed. The models are being compared to eachother to find the most effective method. These models could solve existing problems caused by cancellation, because measures such as overbooking could be used more efficiently. The analysis is based on a real dataset from two portuguese hotels and has almost 120,000 observations and 31 variables.

Questionable research practices have generated considerable recent interest throughout and beyond the scientific community. We subsume such practices involving secret data snooping that influences subsequent statistical inference under the term MESSing (manipulating evidence subject to snooping) and discuss, illustrate and quantify the possibly dramatic effects of several forms of MESSing using an empirical and a simple theoretical example. The empirical example uses numbers from the most popular German lottery, which seem to suggest that 13 is an unlucky number.

Die Einführung der (Spectral-domain-) Optischen Kohärenztomographie (SD-OCT) hat die Vorsorge des Glaukoms revolutioniert. Ihre diagnostischen Möglichkeiten scheinen auch für Patient:innen mit einer herausfordernden Papillenmorphologie im Sinne einer Mikro- oder Makropapille neue Pfade im Rahmen von Prophylaxe und Verlaufsbeurteilung zu eröffnen. Gleichwohl existieren trotz fortgeschrittener Etablierung der Methodik zum Zeitpunkt dieser Studie noch keine festen Referenzwerte oder -bereiche für Makropapillen, da die Mehrzahl der Studien zur Erweiterung der normativen Datengrundlage pri-mär Individuen mit normalgroßen Papillen in die Analysen einbezog. Aus diesem Grunde hat sich die Notwendigkeit einer vergleichenden Analyse hinsichtlich der diagnostischen Wertigkeit der randsaumbasierten Vermessungen im Rahmen der konfokalen Laser-Scanning Tomographie (cSLT) (Neuroretinale Randsaumfläche, DM-RA) sowie der Bruch-Membran-Öffnungsfläche (BMO)-orientierten OCT-Diagnostik (BMO-basierte minimale Randsaumbreite, BMO-MRW und peripapilläre retinale Nervenfaserschichtdicke, RNFL) unter der Berücksichtigung von Makropapillen ergeben. In diese Studie wurden 125 Patient:innen mit Makropapille aus dem Zentrum für Augenheilkunde des Universitätsklinikums Köln eingeschlossen und in drei diagnostische Gruppen eingeteilt: Glaukom (44 Patient:innen), okuläre Hypertension (11 Personen) und ohne Hinweise auf glaukomatöse Veränderungen (70 Pro-band:innen). Die drei im Fokus stehenden morphometrischen Parameter wurden innerhalb der drei Diagnosegruppen ermittelt und bei glaukomatös veränderten Papillen auf Korrelationen untereinander untersucht. Alle Korrelationen waren da-bei signifikant positiv, am stärksten fiel der Zusammenhang zwischen der BMO-MRW und der DM-RA aus (p-Wert < 0,001). Darüber hinaus wurde die Struktur-Funktionsbeziehung der Messparameter ent-lang zweier perimetrischer Stadieneinteilungen untersucht. Dazu wurden diese mit funktionellen Schädigungen als Ergebnis perimetri

Addressing mental health in multimorbidity requires an explanation of a) the genealogy and maintenance of mental health, b) the impact and influence of multimorbidity on mental health, and c) the contextual and intermediate pathways between a) and b). Therefore, the first part of this thesis focuses on theoretical models that give sufficient information on the genealogy and maintenance of mental health in multimorbidity and the impact of multimorbidity on mental health. In the next step, the synthesis c) will result in causal models describing the process between multimorbidity and mental health about the contextual and intermediate factors. In detail, I will point two intermediate factors. First, positive affect, as the most important individual resource, and, second, social support as the most important social resource in preventing deterioration of mental health in multimorbidity.