Wirtschaftswissenschaften
Refine
Year of publication
- 2021 (3) (remove)
Document Type
- Doctoral thesis (3)
Institute
Language
- English (3)
Keywords
- Bayes-Lernen (1)
- Computerlinguistik (1)
- Europe (1)
- Hierarchical bayes model (1)
- IFRS 8 (1)
- International Financial Reporting Standard 8 (1)
- International Financial Reporting Standards (1)
- LDA (1)
- Management Approach (1)
- Management structure (1)
Topic models such as latent Dirichlet allocation (LDA) aim to identify latent topics within text corpora. However, although LDA-type models fall into the category of Natural Language Processing, the actual model input is heavily modified from the original natural language. Among other things, this is typically done by removing specific terms, which arguably might also remove information. In this paper, an extension to LDA is proposed called uLDA, which seeks to incorporate some of these formerly eliminated terms -- namely stop words -- to match natural topics more closely. After developing and evaluating the new extension on established fit measures, uLDA is then tasked with approximating human-perceived topics. For this, a ground truth for topic labels is generated using a human-based experiment. These values are then used as a reference to be matched by the model output. Results show that the new extension outperforms traditional topic models regarding out-of-sample fit across all data sets and regarding human topic approximation for most data sets. These findings demonstrate that the novel extension can extract valuable information from the additional data conveyed by stop words and shows potential for better modeling natural language in the future.
The cumulative dissertation consists of three articles.
The first article analyses the use of vertical manager interlocks for tax avoidance purposes.
The second article studies the effect of tax department organization on tax avoidance and tax risk.
The third article contains an analysis of the effects of tax evasion penalties levied on corporations or managers on aggressive tax avoidance.
Die kumulative Dissertation „The Effectiveness of Segment Disclosures under the Management Approach: Empirical Evidence from Europe” untersucht die Determinanten und ökonomischen Auswirkungen von Segmentberichtsangaben, welche nach International Financial Reporting Standard (IFRS) 8 aufgestellt wurden. Die Dissertation erweitert die bestehende Literatur zur Wirksamkeit des IFRS 8 in Europa und besteht dabei aus drei Beiträgen.
Der erste Beitrag untersucht den Einfluss von Kultur als Determinante für die Quantität und Qualität der Segmentberichtsangaben europäischer Unternehmen. Im Rahmen der empirischen Analyse werden darüber hinaus die ökonomischen Auswirkungen der kulturinduzierten Veröffentlichungsmuster nachgewiesen.
Der zweite Beitrag beantwortet die Forschungsfrage welche Auswirkungen Non-IFRS Segmentdaten auf die Prognosegenauigkeit von Finanzanalysten haben. In der empirischen Analyse wird dieser Wirkungszusammenhang zudem für spezielle Abweichungen von IFRS Rechnungslegungsgrundsätzen untersucht.
Der dritte Beitrag stellt Änderungsvorschläge an IFRS 8 seitens des internationalen Standardsetters vor und untersucht deren potenzielle Auswirkungen aus Perspektive der Jahresabschlussersteller.