1582 research outputs found
Sort by
„Kunstschutz“ as an Alibi: Theodor Wiegand, Georg Karo and the political-military history of German archaeology in Greece and Germany (1909 to 1937). Horizons of expectations of future “Kunstschützer” in occupied Greece (1941 to 1944) and justification strategies after the Second World War
Full text linkDie Abteilung Kunstschutz der deutschen Wehrmacht im besetzten Griechenland (1941-1944) bestand aus wehrpflichtigen deutschen Archäologen. Sie waren zunächst Stipendiaten oder Mitarbeiter des Archäologischen Instituts des Deutschen Reiches (AIDR) unter den Bedingungen des Nationalsozialismus, bevor sie im Zweiten Weltkrieg in der Uniform der Wehrmacht zurückkehrten. Ihre Biografien im Kontext der Abteilung Athen, deren Direktor Georg Karo bis 1936 war, sowie der Zentrale der Instituts, unter dem von 1932 bis 1936 amtierenden Präsidenten Theodor Wiegand, sind ein Untersuchungsgegenstand. Die außenpolitische Legitimation des NS-Regimes durch die Olympischen Spiele und der wichtigste wissenschaftspolitische Erfolg des Institutes, die Wiederaufnahme der Olympiagrabung, die Wiegand und Karo seit 1933 anstrebten und durch ihre politischen Netzwerke 1936 erreichten, werden in der Dissertation in ihrer wechselseitigen Bedingtheit aufgezeigt. Diese Anpassungsleistungen an das NS-Regime prägten den eigenen archäologischen Nachwuchs aber auch die griechische Gesellschaft.
Schutzmaßnahmen waren nur ein kleiner Tätigkeitsbereich der Kunstschützer aber ein wichtiger Teil der Wehrmachtspropaganda. Der Institutspräsident Martin Schede (1937 bis 1945) forderte Mitarbeitern vor allem für zwei AIDR-Projekte an: die Erstellung von Flugbildern von möglichst ganz Griechenland und Ausgrabungen auf Kreta. Bereits diese Zwischenergebnisse berechtigen zu dem Titel „Kunstschutz als Alibi“.
Die Dissertation versucht, die Frage zu beantworten, warum der archäologische Kunstschutz nicht mehr als ein Alibi sein konnte. Dies geschieht vor allem unter Berücksichtigung der politischen aber auch der militärischen Traditionslinien deutscher Archäologie in Griechenland und Deutschland
Exploring the potential of the German Environmental Specimen Bank for retrospective molecular biodiversity monitoring
Full text linkBiodiversity is threatened by a wide range of anthropogenic activities. Monitoring offers critical insights into how and why biodiversity is changing, helping to identify effective measures for maintaining biodiversity and its ecosystem services. However, conventional biodiversity monitoring methods are labor-intensive, and standardized long-term monitoring series are scarce. DNA-based approaches like metabarcoding environmental DNA (eDNA) promise rapid, cost-efficient, and highly resolved community data. At the same time, scientists are looking for alternative data sources that can compensate for the lack of long-term monitoring data to study past biodiversity changes. This work explores the potential of the German Environmental Specimen Bank (ESB), a pollution monitoring archive, which appears particularly promising for retrospective biodiversity monitoring. Biota samples from different ecosystems across the country are collected and archived at an exceptional level of standardization. Sampling species act as natural eDNA samplers, accumulating genetic traces from surrounding organisms. The cryogenic storage at the ESB preserves any eDNA in the samples in its original state. In this thesis, Chapter I serves as an introductory chapter, outlining the general chances and challenges of metabarcoding for assessing biodiversity. Chapter II focuses on primer design and testing the utility of ESB sampling species like mussels and macroalgae for characterizing the surrounding community. Both chapters form the basis for Chapters III to V, which report the use of ESB time series to uncover sample-associated communities and the changes they undergo. Chapter III illustrates the value of these time series by revealing the invasion trajectory of an alien barnacle into German coastal waters and linking the process to climate change. Chapter IV forms the core of this thesis by presenting an expanded measurement of biodiversity change in ESB time series across different taxonomic groups and ecosystem types. Here, a gradual compositional change (turnover) is reported from bacterial, fungal, microeukaryotic, and metazoan communities tending to either spatial homogenization or differentiation. Observed trends are tested for significance using a dynamic model of community ecology based on the equilibrium theory of island biogeography. The model reveals significantly accelerated turnover rates across all taxonomic groups and ecosystems investigated, suggesting a common, anthropogenically induced driver of biodiversity change. Since these analyses most likely include DNA derived from dead as well as from living organisms, Chapter V aims to separate both groups by metabarcoding both DNA and less stable ribosomal RNA from mussel samples. Contrary to the hypothesis, RNA is detectable from both living endobionts and dietary taxa. However, it outcompetes DNA in detecting microeukaryotic biodiversity. In summary, this thesis demonstrates the outstanding potential of archived ESB samples for retrospective biodiversity monitoring, a resource that offers many further untapped opportunities for future biodiversity research at multiple scales
Assessing the Resource and Energy Efficiency of Software and Artificial Intelligence Systems
Full text linkThis dissertation addresses the measurement and evaluation of the energy and resource efficiency of software systems. Studies show that the environmental impact of Information and Communications Technologies (ICT) is steadily increasing and is already estimated to be responsible for 3 % of the total greenhouse gas (GHG) emissions. Although it is the hardware that consumes natural resources and energy through its production, use, and disposal, software controls the hardware and therefore has a considerable influence on the used capacities. Accordingly, it should also be attributed a share of the environmental impact. To address this softwareinduced impact, the focus is on the continued development of a measurement and assessment model for energy and resource-efficient software. Furthermore, measurement and assessment methods from international research and practitioner communities were compared in order to develop a generic reference model for software resource and energy measurements. The next step was to derive a methodology and to define and operationalize criteria for evaluating and improving the environmental impact of software products. In addition, a key objective is to transfer the developed methodology and models to software systems that cause high consumption or offer optimization potential through economies of scale. These include, e. g., Cyber-Physical Systems (CPS) and mobile apps, as well as applications with high demands on computing power or data volumes, such as distributed systems and especially Artificial Intelligence (AI) systems.
In particular, factors influencing the consumption of software along its life cycle are considered. These factors include the location (cloud, edge, embedded) where the computing and storage services are provided, the role of the stakeholders, application scenarios, the configuration of the systems, the used data, its representation and transmission, or the design of the software architecture. Based on existing literature and previous experiments, distinct use cases were selected that address these factors. Comparative use cases include the implementation of a scenario in different programming languages, using varying algorithms, libraries, data structures, protocols, model topologies, hardware and software setups, etc. From the selection, experimental scenarios were devised for the use cases to compare the methods to be analyzed. During their execution, the energy and resource consumption was measured, and the results were assessed. Subtracting baseline measurements of the hardware setup without the software running from the scenario measurements makes the software-induced consumption measurable and thus transparent. Comparing the scenario measurements with each other allows the identification of the more energyefficient setup for the use case and, in turn, the improvement/optimization of the system as a whole. The calculated metrics were then also structured as indicators in a criteria catalog. These indicators represent empirically determinable variables that provide information about a matter that cannot be measured directly, such as the environmental impact of the software. Together with verification criteria that must be complied with and confirmed by the producers of the software, this creates a model with which the comparability of software systems can be established.
The gained knowledge from the experiments and assessments can then be used to forecast and optimize the energy and resource efficiency of software products. This enables developers, but also students, scientists and all other stakeholders involved in the life cycleof software, to continuously monitor and optimize the impact of their software on energy and resource consumption. The developed models, methods, and criteria were evaluated and validated by the scientific community at conferences and workshops. The central outcomes of this thesis, including a measurement reference model and the criteria catalog, were disseminated in academic journals. Furthermore, the transfer to society has been driven forward, e. g., through the publication of two book chapters, the development and presentation of exemplary best practices at developer conferences, collaboration with industry, and the establishment of the eco-label “Blue Angel” for resource and energy-efficient software products. In the long term, the objective is to effect a change in societal attitudes and ultimately to achieve significant resource savings through economies of scale by applying the methods in the development of software in general and AI systems in particular.Diese Arbeit befasst sich mit der Messung und Evaluation der Energie- und Ressourceneffizienz von Softwaresystemen. Studien zeigen, dass die Auswirkungen der Informations- und Kommunikationstechnologien auf die Umwelt stetig zunehmen und bereits heute ca. 3 % der Treibhausgasemissionen verursachen. Zwar ist es die Hardware durch deren Produktion, Nutzung und Entsorgung natürliche Ressourcen und Energie verbraucht werden, aber Software steuert die Hardwarenutzung und hat somit erheblichen Einfluss auf die beanspruchten Kapazitäten. Dementsprechend sollte ihr auch ein Anteil an den verursachten Umweltwirkungen zugeschrieben werden. Um diese, durch die Software hervorgerufenen, Umweltwirkungen zu adressieren steht im Fokus dieser Arbeit die kontinuierliche Entwicklung eines Mess- und Analysemodells für energie- und ressourceneffiziente Software. Darüber hinaus wurden Mess- und Bewertungsmethoden aus internationalen Forschungs- und Praxisgruppen einander gegenübergestellt, um ein allgemeines Referenzmodell für Software Ressourcen- und Energiemessungen zu entwickeln. Es folgt die Ableitung einer Methodik und die Definition und Operationalisierung von Kriterien zur Bewertung und Verbesserung der Umweltwirkungen von Softwareprodukten. Darüber hinaus ist ein wesentliches Ziel die Übertragung der entwickelten Methodik und Modelle auf weitere Softwaresysteme, die hohe Verbräuche hervorrufen oder durch Skaleneffekte großes Optimierungspotential bieten. Dazu zählen beispielsweise cyber-physische Systeme und mobile Apps. Des weiteren wird Software wie verteilte Systeme und insbesondere Anwendungen der Künstlichen Intelligenz (KI) untersucht, die hohe Anforderungen z. B. hinsichtlich der erforderlichen Rechenleistung oder Datenmengen stellen.
Es werden insbesondere Faktoren betrachtet, die den Verbrauch von Softwaresystemen entlang ihres Lebenszyklus beeinflussen. Solche Einflussfaktoren sind beispielsweise der Ort (Cloud, Edge, Embedded) wo die Rechen- und Speicherdienste erbracht werden, die Rolle der Akteure, Anwendungsszenarien, Konfiguration der Systeme, genutzte Daten, deren Darstellung und Übertragung, der Entwurf der Softwarearchitektur etc. Aus diesen möglichen Faktoren wurden, basierend auf bestehender Literatur, sowie vorangegangenen Experimenten möglichst gut abgrenzbare Anwendungsfälle ausgewählt, in denen verschiedene Lösungsansätze zur Zielerreichung eingesetzt und bewertet werden können. Beispiele hierfür sind die Verwendung von verschiedenen Programmiersprachen, Algorithmen, Bibliotheken, Datenstrukturen, Protokollen, Modell-Topologien, Hard- und Softwaresetups. Anhand der Auswahl wurden für die Anwendungsfälle Szenarien jeweils mit den zu untersuchenden Methoden umgesetzt. Während der Ausführung der Szenarien wurden die Energie- und Ressourcenverbräuche gemessen und die Messergebnisse hinsichtlich des Verbrauchs der einzelnen Methoden analysiert. Durch Subtraktion von Baseline-Messungen des Hardware-Setups ohne laufende Software wird der durch die Software hervorgerufene Verbrauch messbar. Dadurch wird Transparenz geschaffen und die ausgewählten Szenarien können miteinander verglichen werden um das für diesen Anwendungsfall energieeffizientere Setup zu identifizieren und so das Gesamtsystem zu verbessern/optimieren. Die berechneten Metriken wurden dann als Indikatoren in einem Kriterienkatalog strukturiert. Diese Indikatoren verkörpern empirisch bestimmbare Größen, die Auskunft über einen nicht direkt messbaren Sachverhalt geben, in diesem Fall die Umweltwirkungen durch die Software. Zusammen mit Nachweiskriterien, die von den Produzenten der Software eingehalten und bestätigt werden müssen entsteht so ein Modell mit dem die Vergleichbarkeit von Softwaresystemen hergestellt wird.
Schließlich dienen die in den Versuchen gewonnenen Erkenntnisse zur Prognose und Optimierung der Energie- und Ressourceneffizienz der Softwareprodukte. Durch die entwickelten Modelle sollen insbesondere Entwickler:innen, aber auch Studierende, Wissenschaftler:innen und allen anderen am Lebenszyklus einer Software beteiligten Stakeholder in die Lage versetzt werden die Auswirkungen ihrer Software auf den Energie- und Ressourcenverbrauch kontinuierlich zu überwachen und optimiern. Die entwickelten Modelle, Methoden und Kriterien wurden durch die wissenschaftliche Community auf Konferenzen und Workshops evaluiert und validiert. Zentrale Ergebnisse dieser Arbeit wie ein Mess-Referenzmodell und der Kriterienkatalog wurden in Journalen verbreitet. Darüber hinaus wird der Transfer in die Gesellschaft vorabgetrieben, z. B. durch zwei Buchkapitel, die Erarbeitung und Vorstellung von Best Practice Beispielen auf Entwicklerkonferenzen, der Kooperation mit der Industrie oder durch die Entwicklung von Labels, wie dem "Blauen Engel" für ressourcen- und energieeffiziente Softwareprodukte. Langfristig ist das Ziel zu einem Umdenken in der Gesellschaft anzuregen und durch Skaleneffekte schließlich starke Ressourceneinsparungen durch die Anwendung der Methoden in der Entwicklung von Software und insbesondere KI-Systemen zu erreichen
Dèi e Zangrèi: La lingua ferita, l'identità negata. Gli Elleni di Calabria e i Lombardi di Sicilia
Full text linkNel libro Dèi e Zangrèi il professor Pasquale Casile scandaglia con mirabile precisione scientifica e in tutta profondità gli abissi della memoria linguistica dei Greci di Calabria. Fornisce risposte valide ai quesiti: Chi sono gli Zangrèi? Sono gli ultimi Dionisiaci della storia e gli eredi diretti delle comunità orfico-pitagoriche della Magna Grecia? E perché vengono così chiamati anche i Catari di Sicilia
Essays on Financial Market Regimes
Full text linkThis dissertation examines the relevance of regimes for stock markets. In three research articles, we cover the identification and predictability of regimes and their relationships to macroeconomic and financial variables in the United States.
The initial two chapters contribute to the debate on the predictability of stock markets. While various approaches can demonstrate in-sample predictability, their predictive power diminishes substantially in out-of-sample studies. Parameter instability and model uncertainty are the primary challenges. However, certain methods have demonstrated efficacy in addressing these issues. In Chapter 1 and 2, we present frameworks that combine these methods meaningfully. Chapter 3 focuses on the role of regimes in explaining macro-financial relationships and examines the state-dependent effects of macroeconomic expectations on cross-sectional stock returns. Although it is common to capture the variation in stock returns using factor models, their macroeconomic risk sources are unclear. According to macro-financial asset pricing, expectations about state variables may be viable candidates to explain these sources. We examine their usefulness in explaining factor premia and assess their suitability for pricing stock portfolios.
In summary, this dissertation improves our understanding of stock market regimes in three ways. First, we show that it is worthwhile to exploit the regime dependence of stock markets. Markov-switching models and their extensions are valuable tools for filtering the stock market dynamics and identifying and predicting regimes in real-time. Moreover, accounting for regime-dependent relationships helps to examine the dynamic impact of macroeconomic shocks on stock returns. Second, we emphasize the usefulness of macro-financial variables for the stock market. Regime identification and forecasting benefit from their inclusion. This is particularly true in periods of high uncertainty when information processing in financial markets is less efficient. Finally, we recommend to address parameter instability, estimation risk, and model uncertainty in empirical models. Because it is difficult to find a single approach that meets all of these challenges simultaneously, it is advisable to combine appropriate methods in a meaningful way. The framework should be as complex as necessary but as parsimonious as possible to mitigate additional estimation risk. This is especially recommended when working with financial market data with a typically low signal-to-noise ratio
Erstellen von 3D-Objekten aus Fotos und deren lagegenaue Darstellung mithilfe von mobiler AR
Full text linkIn dieser Dissertation wird der Workflow der Erstellung einer Augmented Reality App für das Projekt „ARmob” auf Androidgeräten beschrieben. Diese App positioniert durch SfM-Technik erstellte, nach dem neuesten Stand der Forschung rekonstruierte 3D-Objekte an ihren ursprünglichen Standort in der Realität. Die virtuellen Objekte werden jeweils vom Standpunkt und Blickwinkel des Betrachters passend in die reale Welt eingeblendet, so dass der Eindruck entsteht, die Objekte seien Teil der Realität. Die lagegenaue Darstellung ist abhängig von der Satellitenerreichbarkeit der GNSS und der Genauigkeit der weiteren Sensoren. Die App soll als Grundlage und Framework für weitere Apps zur Erforschung der Raumwahrnehmung im Bereich der Kartographie dienen.This dissertation describes the workflow for the creation of an augmented reality app for the “ARmob” project on Android devices. This app positions 3D objects created with SfM technique at their original location in reality. The virtual objects are displayed from the viewer's point of view and perspective of the real world, giving him or her the impression that the objects are part of reality. The exact positional representation depends on the satellite accessibility of the GNSS and the accuracy of the other sensors. The app is intended to serve as a basis and framework for further apps for researching spatial perception in the field of cartography
Optimization for Fair Classification Methods in Heterogeneous Data
Full text linkEnsuring fairness in machine learning models is crucial for ethical and unbiased automated decision-making. Classifications from fair machine learning models should not discriminate against sensitive variables such as sexual orientation and ethnicity. However, achieving fairness is complicated by biases inherent in training data, particularly when data is collected through group sampling, like stratified or cluster sampling as often occurs in social surveys. Unlike the standard assumption of independent observations in machine learning, clustered data introduces correlations that can amplify biases, especially when cluster assignment is linked to the target variable.
To address these challenges, this cumulative thesis focuses on developing methods to mitigate unfairness in machine learning models. We propose a fair mixed effects support vector machine algorithm, a Cluster-Regularized Logistic Regression and a fair Generalized Linear Mixed Model based on boosting, all of them are capable of handling both grouped data and fairness constraints simultaneously. Additionally, we introduce a Julia package, FairML.jl, which provides a comprehensive framework for addressing fairness issues. This package offers a preprocessing technique, based on resampling methods, to mitigate biases in the data, as well as a post-processing method, that seeks for a optimal cut-off selection.
To improve fairness in classifications both processes can be incorporated in any classification method available in the MLJ.jl package. Furthermore, FairML.jl incorporates in-processing approaches, such as optimization-based techniques for logistic regression and support vector machine, to directly address fairness during model training in regular and mixed models.
By accounting for data complexities and implementing various fairness-enhancing strategies, our work aims to contribute to the development of more equitable and reliable machine learning models
An Explainable Artificial Intelligence Text Classifier for Suicidality Prediction in Youth Crisis Text Line Users: Development and Validation Study
Full text linkBackground: Suicide represents a critical public health concern, and machine learning (ML) models offer the potential for identifying at-risk individuals. Recent studies using benchmark datasets and real-world social media data have demonstrated the capability of pretrained large language models in predicting suicidal ideation and behaviors (SIB) in speech and text.
Objective: This study aimed to (1) develop and implement ML methods for predicting SIBs in a real-world crisis helpline dataset, using transformer-based pretrained models as a foundation; (2) evaluate, cross-validate, and benchmark the model against traditional text classification approaches; and (3) train an explainable model to highlight relevant risk-associated features.
Methods: We analyzed chat protocols from adolescents and young adults (aged 14-25 years) seeking assistance from a German crisis helpline. An ML model was developed using a transformer-based language model architecture with pretrained weights and long short-term memory layers. The model predicted suicidal ideation (SI) and advanced suicidal engagement (ASE), as indicated by composite Columbia-Suicide Severity Rating Scale scores. We compared model performance against a classical word-vector-based ML model. We subsequently computed discrimination, calibration, clinical utility, and explainability information using a Shapley Additive Explanations value-based post hoc estimation model.
Results: The dataset comprised 1348 help-seeking encounters (1011 for training and 337 for testing). The transformer-based classifier achieved a macroaveraged area under the curve (AUC) receiver operating characteristic (ROC) of 0.89 (95% CI 0.81-0.91) and an overall accuracy of 0.79 (95% CI 0.73-0.99). This performance surpassed the word-vector-based baseline model (AUC-ROC=0.77, 95% CI 0.64-0.90; accuracy=0.61, 95% CI 0.61-0.80). The transformer model demonstrated excellent prediction for nonsuicidal sessions (AUC-ROC=0.96, 95% CI 0.96-0.99) and good prediction for SI and ASE, with AUC-ROCs of 0.85 (95% CI 0.97-0.86) and 0.87 (95% CI 0.81-0.88), respectively. The Brier Skill Score indicated a 44% improvement in classification performance over the baseline model. The Shapley Additive Explanations model identified language features predictive of SIBs, including self-reference, negation, expressions of low self-esteem, and absolutist language.
Conclusions: Neural networks using large language model–based transfer learning can accurately identify SI and ASE. The post hoc explainer model revealed language features associated with SI and ASE. Such models may potentially support clinical decision-making in suicide prevention services. Future research should explore multimodal input features and temporal aspects of suicide risk
Exploring Biases of Large Language Models in the Field of Mental Health: Comparative Questionnaire Study of the Effect of Gender and Sexual Orientation in Anorexia Nervosa and Bulimia Nervosa Case Vignettes
Full text linkBackground: Large language models (LLMs) are increasingly used in mental health, showing promise in assessing disorders. However, concerns exist regarding their accuracy, reliability, and fairness. Societal biases and underrepresentation of certain populations may impact LLMs. Because LLMs are already used for clinical practice, including decision support, it is important to investigate potential biases to ensure a responsible use of LLMs. Anorexia nervosa (AN) and bulimia nervosa (BN) show a lifetime prevalence of 1%-2%, affecting more women than men. Among men, homosexual men face a higher risk of eating disorders (EDs) than heterosexual men. However, men are underrepresented in ED research, and studies on gender, sexual orientation, and their impact on AN and BN prevalence, symptoms, and treatment outcomes remain limited.
Objectives: We aimed to estimate the presence and size of bias related to gender and sexual orientation produced by a common LLM as well as a smaller LLM specifically trained for mental health analyses, exemplified in the context of ED symptomatology and health-related quality of life (HRQoL) of patients with AN or BN.
Methods: We extracted 30 case vignettes (22 AN and 8 BN) from scientific papers. We adapted each vignette to create 4 versions, describing a female versus male patient living with their female versus male partner (2 × 2 design), yielding 120 vignettes. We then fed each vignette into ChatGPT-4 and to “MentaLLaMA” based on the Large Language Model Meta AI (LLaMA) architecture thrice with the instruction to evaluate them by providing responses to 2 psychometric instruments, the RAND-36 questionnaire assessing HRQoL and the eating disorder examination questionnaire. With the resulting LLM-generated scores, we calculated multilevel models with a random intercept for gender and sexual orientation (accounting for within-vignette variance), nested in vignettes (accounting for between-vignette variance).
Results: In ChatGPT-4, the multilevel model with 360 observations indicated a significant association with gender for the RAND-36 mental composite summary (conditional means: 12.8 for male and 15.1 for female cases; 95% CI of the effect –6.15 to -0.35; P=.04) but neither with sexual orientation (P=.71) nor with an interaction effect (P=.37). We found no indications for main effects of gender (conditional means: 5.65 for male and 5.61 for female cases; 95% CI –0.10 to 0.14; P=.88), sexual orientation (conditional means: 5.63 for heterosexual and 5.62 for homosexual cases; 95% CI –0.14 to 0.09; P=.67), or for an interaction effect (P=.61, 95% CI –0.11 to 0.19) for the eating disorder examination questionnaire overall score (conditional means 5.59-5.65 95% CIs 5.45 to 5.7). MentaLLaMA did not yield reliable results.
Conclusions: LLM-generated mental HRQoL estimates for AN and BN case vignettes may be biased by gender, with male cases scoring lower despite no real-world evidence supporting this pattern. This highlights the risk of bias in generative artificial intelligence in the field of mental health. Understanding and mitigating biases related to gender and other factors, such as ethnicity, and socioeconomic status are crucial for responsible use in diagnostics and treatment recommendations
Large Players, Information and Coordination
Full text linkIn this dissertation, I analyze how large players in financial markets exert influence on smaller players and how this affects the decisions of the large ones. I focus on how the large players process information in an uncertain environment, form expectations and communicate these to smaller players through their actions. I examine these relationships empirically in the foreign exchange market and in the context of a game-theoretic model of an investment project.
In Chapter 2, I investigate the relationship between the foreign exchange trading activity of large US-based market participants and the volatility of the nominal spot exchange rate. Using a novel dataset, I utilize the weekly growth rate of aggregate foreign currency positions of major market participants to proxy trading activity in the foreign exchange market. By estimating the heterogeneous autoregressive model of realized volatility (HAR-RV), I find evidence of a positive relationship between trading activity and volatility, which is mainly driven by unexpected changes in trading activity and is asymmetric for some of the currencies considered. My results contribute to the understanding of the drivers of exchange rate volatility and the role of large players in the flow of information in financial markets.
In Chapters 3 and 4, I consider a sequential global game of an investment project to examine how a large creditor influences the decisions of small creditors with her lending decision. I pay particular attention to the timing of the large player’s decision, i.e. whether she makes her decision to roll over a credit before or after the small players. I show that she faces a trade-off between signaling to and learning from small creditors. By being a focal point for coordination, her actions have a substantial impact on the probability of coordination failure and the failure of the investment project. I investigate the sensitivity of the equilibrium by comparing settings with perfect and imperfect learning. The results highlight the importance of signaling and provide a new perspective on the idea of catalytic finance and the influence of a lender-of-last-resort in self-fulfilling debt crises.In dieser Dissertation untersuche ich, wie große Akteure auf den Finanzmärkten Einfluss auf kleinere Akteure ausüben und wie sich dies auf die Entscheidungen der großen Akteure auswirkt. Ein besonderer Fokus liegt dabei darauf, wie die großen Akteure in einem unsicheren Umfeld Informationen verarbeiten, Erwartungen bilden und diese durch ihr Handeln an kleinere Akteure kommunizieren. Ich untersuche diese Zusammenhänge empirisch am Devisenmarkt und im Rahmen eines spieltheoretischen Modells eines Investitionsprojekts.
In Kapitel 2 untersuche ich die Beziehung zwischen der Devisenhandelsaktivität großer Marktteilnehmer in den USA und der Volatilität des nominalen Devisenkassakurses. Unter Verwendung eines neuartigen Datensatzes nutze ich die wöchentliche Wachstumsrate der aggregierten Fremdwährungspositionen großer Marktteilnehmer, um die Handelsaktivität auf dem Devisenmarkt zu ermitteln. Durch Schätzung des Heterogenen Autoregressiven Modells der Realisierten Volatilität (HAR-RV) finde ich Belege für eine positive Beziehung zwischen Handelsaktivität und Volatilität, die hauptsächlich durch unerwartete Veränderungen der Handelsaktivität bedingt ist und für einige der betrachteten Währungen asymmetrisch ist. Meine Ergebnisse tragen zum Verständnis der Triebkräfte der Wechselkursvolatilität und der Rolle der großen Akteure für den Informationsfluss auf Finanzmärkten bei.
In den Kapiteln 3 und 4 betrachte ich ein sequentielles globales Spiel eines Investitionsprojekts, um zu untersuchen, wie ein großer Kreditgeber mit seiner Kreditentscheidung die Entscheidungen kleiner Kreditgeber beeinflusst. Besonderes Augenmerk lege ich dabei auf den Zeitpunkt der Entscheidung des großen Akteurs, d.h. ob er seine Entscheidung, einen Kredit zu verlängern, vor oder nach den kleinen Akteuren trifft. Ich zeige, dass er einen Kompromiss zwischen der Signalwirkung für die kleinen Gläubiger und dem Lernen von ihnen eingeht. Da seine Entscheidung einen Orientierungspunkt für die Koordinierung darstellt, hat diese einen erheblichen Einfluss auf die Wahrscheinlichkeit eines Koordinationsversagens und des Scheiterns des Investitionsprojekts. Ich untersuche die Empfindlichkeit des Gleichgewichts, indem ich Szenarien mit perfektem und unvollkommenem Lernen vergleiche. Die Ergebnisse unterstreichen die Bedeutung von Signalen und bieten eine neue Perspektive auf die Idee der katalytischen Finanzierung und den Einfluss eines Lender-of-Last-Resort in sich selbst erfüllenden Schuldenkrisen