Refine
Year of publication
Document Type
- Doctoral Thesis (61)
- Article (3)
- Report (1)
- Working Paper (1)
Language
- German (36)
- English (29)
- Multiple languages (1)
Keywords
- Deutschland (7)
- China (3)
- Entrepreneurship (3)
- Nachhaltigkeit (3)
- Simulation (3)
- Amtliche Statistik (2)
- COVID-19 (2)
- Crowdfunding (2)
- Diversifikation (2)
- European Union (2)
Institute
- Wirtschaftswissenschaften (66) (remove)
Das Ziel dynamischer Mikrosimulationen ist es, die Entwicklung von Systemen über das Verhalten der einzelnen enthaltenen Bestandteile zu simulieren, um umfassende szenariobasierte Analysen zu ermöglichen. Im Bereich der Wirtschafts- und Sozialwissenschaften wird der Fokus üblicherweise auf Populationen bestehend aus Personen und Haushalten gelegt. Da politische und wirtschaftliche Entscheidungsprozesse meist auf lokaler Ebene getroffen werden, bedarf es zudem kleinräumiger Informationen, um gezielte Handlungsempfehlungen ableiten zu können. Das stellt Forschende wiederum vor große Herausforderungen im Erstellungsprozess regionalisierter Simulationsmodelle. Dieser Prozess reicht von der Generierung geeigneter Ausgangsdatensätze über die Erfassung und Umsetzung der dynamischen Komponenten bis hin zur Auswertung der Ergebnisse und Quantifizierung von Unsicherheiten. Im Rahmen dieser Arbeit werden ausgewählte Komponenten, die für regionalisierte Mikrosimulationen von besonderer Relevanz sind, beschrieben und systematisch analysiert.
Zunächst werden in Kapitel 2 theoretische und methodische Aspekte von Mikrosimulationen vorgestellt, um einen umfassenden Überblick über verschiedene Arten und Möglichkeiten der Umsetzung dynamischer Modellierungen zu geben. Im Fokus stehen dabei die Grundlagen der Erfassung und Simulation von Zuständen und Zustandsänderungen sowie die damit verbundenen strukturellen Aspekte im Simulationsprozess.
Sowohl für die Simulation von Zustandsänderungen als auch für die Erweiterung der Datenbasis werden primär logistische Regressionsmodelle zur Erfassung und anschließenden wahrscheinlichkeitsbasierten Vorhersage der Bevölkerungsstrukturen auf Mikroebene herangezogen. Die Schätzung beruht insbesondere auf Stichprobendaten, die in der Regel neben einem eingeschränktem Stichprobenumfang keine oder nur unzureichende regionale Differenzierungen zulassen. Daher können bei der Vorhersage von Wahrscheinlichkeiten erhebliche Differenzen zu bekannten Totalwerten entstehen. Um eine Harmonisierung mit den Totalwerten zu erhalten, lassen sich Methoden zur Anpassung von Wahrscheinlichkeiten – sogenannte Alignmentmethoden – anwenden. In der Literatur werden zwar unterschiedliche Möglichkeiten beschrieben, über die Auswirkungen dieser Verfahren auf die Güte der Modelle ist jedoch kaum etwas bekannt. Zur Beurteilung verschiedener Techniken werden diese im Rahmen von Kapitel 3 in umfassenden Simulationsstudien unter verschiedenen Szenarien umgesetzt. Hierbei kann gezeigt werden, dass durch die Einbindung zusätzlicher Informationen im Modellierungsprozess deutliche Verbesserungen sowohl bei der Schätzung der Parameter als auch bei der Vorhersage der Wahrscheinlichkeiten erzielt werden können. Zudem lassen sich dadurch auch bei fehlenden regionalen Identifikatoren in den Modellierungsdaten kleinräumige Wahrscheinlichkeiten erzeugen. Insbesondere die Maximierung der Likelihood des zugrundeliegenden Regressionsmodells unter der Nebenbedingung, dass die bekannten Totalwerte eingehalten werden, weist in allen Simulationsstudien überaus gute Ergebnisse auf.
Als eine der einflussreichsten Komponenten in regionalisierten Mikrosimulationen erweist sich die Umsetzung regionaler Mobilität. Gleichzeitig finden Wanderungen in vielen Mikrosimulationsmodellen keine oder nur unzureichende Beachtung. Durch den unmittelbaren Einfluss auf die gesamte Bevölkerungsstruktur führt ein Ignorieren jedoch bereits bei einem kurzen Simulationshorizont zu starken Verzerrungen. Während für globale Modelle die Integration von Wanderungsbewegungen über Landesgrenzen ausreicht, müssen in regionalisierten Modellen auch Binnenwanderungsbewegungen möglichst umfassend nachgebildet werden. Zu diesem Zweck werden in Kapitel 4 Konzepte für Wanderungsmodule erstellt, die zum einen eine unabhängige Simulation auf regionalen Subpopulationen und zum anderen eine umfassende Nachbildung von Wanderungsbewegungen innerhalb der gesamten Population zulassen. Um eine Berücksichtigung von Haushaltsstrukturen zu ermöglichen und die Plausibilität der Daten zu gewährleisten, wird ein Algorithmus zur Kalibrierung von Haushaltswahrscheinlichkeiten vorgeschlagen, der die Einhaltung von Benchmarks auf Individualebene ermöglicht. Über die retrospektive Evaluation der simulierten Migrationsbewegungen wird die Funktionalität der Wanderdungskonzepte verdeutlicht. Darüber hinaus werden über die Fortschreibung der Population in zukünftige Perioden divergente Entwicklungen der Einwohnerzahlen durch verschiedene Konzepte der Wanderungen analysiert.
Eine besondere Herausforderung in dynamischen Mikrosimulationen stellt die Erfassung von Unsicherheiten dar. Durch die Komplexität der gesamten Struktur und die Heterogenität der Komponenten ist die Anwendung klassischer Methoden zur Messung von Unsicherheiten oft nicht mehr möglich. Zur Quantifizierung verschiedener Einflussfaktoren werden in Kapitel 5 varianzbasierte Sensitivitätsanalysen vorgeschlagen, die aufgrund ihrer enormen Flexibilität auch direkte Vergleiche zwischen unterschiedlichsten Komponenten ermöglichen. Dabei erweisen sich Sensitivitätsanalysen nicht nur für die Erfassung von Unsicherheiten, sondern auch für die direkte Analyse verschiedener Szenarien, insbesondere zur Evaluation gemeinsamer Effekte, als überaus geeignet. In Simulationsstudien wird die Anwendung im konkreten Kontext dynamischer Modelle veranschaulicht. Dadurch wird deutlich, dass zum einen große Unterschiede hinsichtlich verschiedener Zielwerte und Simulationsperioden auftreten, zum anderen aber auch immer der Grad an regionaler Differenzierung berücksichtigt werden muss.
Kapitel 6 fasst die Erkenntnisse der vorliegenden Arbeit zusammen und gibt einen Ausblick auf zukünftige Forschungspotentiale.
Non-probability sampling is a topic of growing relevance, especially due to its occurrence in the context of new emerging data sources like web surveys and Big Data.
This thesis addresses statistical challenges arising from non-probability samples, where unknown or uncontrolled sampling mechanisms raise concerns in terms of data quality and representativity.
Various methods to quantify and reduce the potential selectivity and biases of non-probability samples in estimation and inference are discussed. The thesis introduces new forms of prediction and weighting methods, namely
a) semi-parametric artificial neural networks (ANNs) that integrate B-spline layers with optimal knot positioning in the general structure and fitting procedure of artificial neural networks, and
b) calibrated semi-parametric ANNs that determine weights for non-probability samples by integrating an ANN as response model with calibration constraints for totals, covariances and correlations.
Custom-made computational implementations are developed for fitting (calibrated) semi-parametric ANNs by means of stochastic gradient descent, BFGS and sequential quadratic programming algorithms.
The performance of all the discussed methods is evaluated and compared for a bandwidth of non-probability sampling scenarios in a Monte Carlo simulation study as well as an application to a real non-probability sample, the WageIndicator web survey.
Potentials and limitations of the different methods for dealing with the challenges of non-probability sampling under various circumstances are highlighted. It is shown that the best strategy for using non-probability samples heavily depends on the particular selection mechanism, research interest and available auxiliary information.
Nevertheless, the findings show that existing as well as newly proposed methods can be used to ease or even fully counterbalance the issues of non-probability samples and highlight the conditions under which this is possible.
The Belt and Road Initiative (BRI) has had a significant impact on China in political, economic, and cultural terms. This study focuses on the cultural domain, especially on scholarship students from the countries that signed bilateral cooperation agreements with China under the BRI. Using an integrated approach combining the difference-in-differences method and the gravity model, we explore the correlation between the BRI and the increasing number of international scholarship students funded by the Chinese government, as well as the determinants of students' decision to study in China. The panel data from 2010 to 2018 show that the launch of BRI has had a positive impact on the number of scholarship students from BRI countries. The number of scholarship recipients from non-BRI countries also increased, but at a much slower rate than those from BRI countries. The sole exception is the United States, which has trended downward for both state-funded and self-funded students.
The outbreak of the COVID-19 pandemic has also led to many conspiracy theories. While the origin of the pandemic in China led some, including former US president Donald Trump, to dub the pathogen “Chinese virus” and to support anti-Chinese conspiracy narratives, it caused Chinese state officials to openly support anti-US conspiracy theories about the “true” origin of the virus. In this article, we study whether nationalism, or more precisely uncritical patriotism, is related to belief in conspiracy theories among normal people. We hypothesize based on group identity theory and motivated reasoning that for the particular case of conspiracy theories related to the origin of COVID-19, such a relation should be stronger for Chinese than for Germans. To test this hypothesis, we use survey data from Germany and China, including data from the Chinese community in Germany. We also look at relations to other factors, in particular media consumption and xenophobia.
We study planned changes in protective routines after the COVID-19 pandemic: in a survey in Germany among >650 respondents, we find that the majority plans to use face masks in certain situations even after the end of the pandemic. We observe that this willingness is strongly related to the perception that there is something to be learned from East Asians’ handling of pandemics, even when controlling for perceived protection by wearing masks. Given strong empirical evidence that face masks help prevent the spread of respiratory diseases and given the considerable estimated health and economic costs of such diseases even pre-Corona, this would be a very positive side effect of the current crisis.
Retirement, fertility and sexuality are three key life stage events that are embedded in the framework of population economics in this dissertation. Each topic implies economic relevance. As retirement entry shifts labour supply of experienced workers to zero, this issue is particularly relevant for employers, retirees themselves as well as policymakers who are in charge of the design of the pension system. Giving birth has comprehensive economic relevance for women. Parental leave and subsequent part-time work lead to a direct loss of income. Lower levels of employment, work experience, training and career opportunities result in indirect income losses. Sexuality has decisive influence on the quality of partnerships, subjective well-being and happiness. Well-being and happiness, in turn, are significant key determinants not only in private life but also in the work domain, for example in the area of job performance. Furthermore, partnership quality determines the duration of a partnership. And in general, partnerships enable the pooling of (financial) resources - compared to being single. The contribution of this dissertation emerges from the integration of social and psychological concepts into economic analysis as well as the application of economic theory in non-standard economic research topics. The results of the three chapters show that the multidisciplinary approach yields better prediction of human behaviour than the single disciplines on their own. The results in the first chapter show that both interpersonal conflict with superiors and the individual’s health status play a significant role in retirement decisions. The chapter further contributes to existing literature by showing the moderating role of health within the retirement decision-making: On the one hand, all employees are more likely to retire when they are having conflicts with their superior. On the other hand, among healthy employees, the same conflict raises retirement intentions even more. That means good health is a necessary, but not a sufficient condition for continued working. It may be that conflicts with superiors raise retirement intentions more if the worker is healthy. The key findings of the second chapter reveal significant influence of religion on contraceptive and fertility-related decisions. A large part of research on religion and fertility is originated in evidence from the US. This chapter contrasts evidence from Germany. Additionally, the chapter contributes by integrating miscarriages and abortions, rather than limiting the analysis to births and it gains from rich prospective data on fertility biography of women. The third chapter provides theoretical insights on how to incorporate psychological variables into an economic framework which aims to analyse sexual well-being. According to this theory, personality may play a dual role by shaping a person’s preferences for sex as well as the person’s behaviour in a sexual relationship. Results of econometric analysis reveal detrimental effects of neuroticism on sexual well-being while conscientiousness seems to create a win-win situation for a couple. Extraversions and Openness have ambiguous effects on romantic relationships by enhancing sexual well-being on the one hand but raising commitment problems on the other. Agreeable persons seem to gain sexual satisfaction even if they perform worse in sexual communication.
Mittels Querschnittserhebungen ist es möglich Populationsparameter zu einem bestimmten Zeitpunkt zu schätzen. Jedoch ist meist die Veränderung von Populationsparametern von besonderem Interesse. So ist es zur Evaluation von politischen Zielvorgaben erforderlich die Veränderung von Indikatoren, wie Armutsmaßen, über die Zeit zu verfolgen. Um zu testen ob eine gemessene Veränderung sich signifikant von Null unterscheidet bedarf es einer Varianzschätzung für Veränderungen von Querschnitten. In diesem Zusammenhang ergeben sich oft zwei Probleme; Zum einen sind die relevanten Statistiken meist nicht-linear und zum anderen basieren die untersuchten Querschnittserhebungen auf Stichproben die nicht unabhängig voneinander gezogen wurden. Ziel der vorliegenden Dissertation ist es einen theoretischen Rahmen zur Herleitung und Schätzung der Varianz einer geschätzten Veränderung von nicht-linearen Statistiken zu geben. Hierzu werden die Eigenschaften von Stichprobendesigns erarbeitetet, die zur Koordination von Stichprobenziehungen in einer zeitlichen Abfolge verwendet werden. Insbesondere werden Ziehungsalgorithmen zur Koordination von Stichproben vorgestellt, erarbeitet und deren Eigenschaften beschrieben. Die Problematik der Varianzschätzung im Querschnitt für nicht-lineare Schätzer bei komplexen Stichprobendesigns wird ebenfalls behandelt. Schließlich wird ein allgemeiner Ansatz zur Schätzung von Veränderungen aufgezeigt und es werden Varianzschätzer für die Veränderung von Querschnittschätzern basierend auf koordinierten Querschnittstichproben untersucht. Insbesondere dem Fall einer sich über die Zeit verändernden Population wird eine besondere Bedeutung im Rahmen der Arbeit beigemessen, da diese im Anwendungsfall die Regel darstellen.
Surveys are commonly tailored to produce estimates of aggregate statistics with a desired level of precision. This may lead to very small sample sizes for subpopulations of interest, defined geographically or by content, which are not incorporated into the survey design. We refer to subpopulations where the sample size is too small to provide direct estimates with adequate precision as small areas or small domains. Despite the small sample sizes, reliable small area estimates are needed for economic and political decision making. Hence, model-based estimation techniques are used which increase the effective sample size by borrowing strength from other areas to provide accurate information for small areas. The paragraph above introduced small area estimation as a field of survey statistics where two conflicting philosophies of statistical inference meet: the design-based and the model-based approach. While the first approach is well suited for the precise estimation of aggregate statistics, the latter approach furnishes reliable small area estimates. In most applications, estimates for both large and small domains based on the same sample are needed. This poses a challenge to the survey planner, as the sampling design has to reflect different and potentially conflicting requirements simultaneously. In order to enable efficient design-based estimates for large domains, the sampling design should incorporate information related to the variables of interest. This may be achieved using stratification or sampling with unequal probabilities. Many model-based small area techniques require an ignorable sampling design such that after conditioning on the covariates the variable of interest does not contain further information about the sample membership. If this condition is not fulfilled, biased model-based estimates may result, as the model which holds for the sample is different from the one valid for the population. Hence, an optimisation of the sampling design without investigating the implications for model-based approaches will not be sufficient. Analogously, disregarding the design altogether and focussing only on the model is prone to failure as well. Instead, a profound knowledge of the interplay between the sample design and statistical modelling is a prerequisite for implementing an effective small area estimation strategy. In this work, we concentrate on two approaches to address this conflict. Our first approach takes the sampling design as given and can be used after the sample has been collected. It amounts to incorporate the survey design into the small area model to avoid biases stemming from informative sampling. Thus, once a model is validated for the sample, we know that it holds for the population as well. We derive such a procedure under a lognormal mixed model, which is a popular choice when the support of the dependent variable is limited to positive values. Besides, we propose a three pillar strategy to select the additional variable accounting for the design, based on a graphical examination of the relationship, a comparison of the predictive accuracy of the choices and a check regarding the normality assumptions.rnrnOur second approach to deal with the conflict is based on the notion that the design should allow applying a wide variety of analyses using the sample data. Thus, if the use of model-based estimation strategies can be anticipated before the sample is drawn, this should be reflected in the design. The same applies for the estimation of national statistics using design-based approaches. Therefore, we propose to construct the design such that the sampling mechanism is non-informative but allows for precise design-based estimates at an aggregate level.
A phenomenon of recent decades is that digital marketplaces on the Internet are establishing themselves for a wide variety of products and services. Recently, it has become possible for private individuals to invest in young and innovative companies (so-called "start-ups"). Via Internet portals, potential investors can examine various start-ups and then directly invest in their chosen start-up. In return, investors receive a share in the firm- profit, while companies can use the raised capital to finance their projects. This new way of financing is called "Equity Crowdfunding" (ECF) or "Crowdinvesting". The aim of this dissertation is to provide empirical findings about the characteristics of ECF. In particular, the question of whether ECF is able to overcome geographic barriers, the interdependence of ECF and capital structure, and the risk of failure for funded start-ups and their chances of receiving follow-up funding by venture capitalists or business angels will be analyzed. The results of the first part of this dissertation show that investors in ECF prefer local companies. In particular, investors who invest larger amounts have a stronger tendency to invest in local start-ups. The second part of the dissertation provides first indications of the interdependencies between capital structure and ECF. The analysis makes clear that the capital structure is not a determinant for undertaking an ECF campaign. The third part of the dissertation analyzes the success of companies financed by ECF in a country comparison. The results show that after a successful ECF campaign German companies have a higher chance of receiving follow-up funding by venture capitalists compared to British companies. The probability of survival, however, is slightly lower for German companies. The results provide relevant implications for theory and practice. The existing literature in the area of entrepreneurial finance will be extended by insights into investor behavior, additions to the capital structure theory and a country comparison in ECF. In addition, implications are provided for various actors in practice.
Entrepreneurship is a process of discovering and exploiting opportunities, during which two crucial milestones emerge: in the very beginning when entrepreneurs start their businesses, and in the end when they determine the future of the business. This dissertation examines the establishment and exit of newly created as well as of acquired firms, in particular the behavior and performance of entrepreneurs at these two important stages of entrepreneurship. The first part of the dissertation investigates the impact of characteristics at the individual and at the firm level on an entrepreneur- selection of entry modes across new venture start-up and business takeover. The second part of the dissertation compares firm performance across different entrepreneurship entry modes and then examines management succession issues that family firm owners have to confront. This study has four main findings. First, previous work experience in small firms, same sector experience, and management experience affect an entrepreneur- choice of entry modes. Second, the choice of entry mode for hybrid entrepreneurs is associated with their characteristics, such as occupational experience, level of education, and gender, as well as with the characteristics of their firms, such as location. Third, business takeovers survive longer than new venture start-ups, and both entry modes have different survival determinants. Fourth, the family firm- decision of recruiting a family or a nonfamily manager is not only determined by a manager- abilities, but also by the relationship between the firm- economic and non-economic goals and the measurability of these goals. The findings of this study extend our knowledge on entrepreneurship entry modes by showing that new venture start-ups and business takeovers are two distinct entrepreneurship entry modes in terms of their founders" profiles, their survival rates and survival determinants. Moreover, this study contributes to the literature on top management hiring in family firms: it establishes family firm- non-economic goals as another factor that impacts the family firm- hiring decision between a family and a nonfamily manager.