ZU|Repositorium – der Hochschul- und Dokumentenschriftenserver der Zeppelin Universität
Not a member yet
12136 research outputs found
Sort by
Organising the Popular? Organisation, Political Inclusion, and the Challenges of Populism
No full textAre non-democracies taking over? Autocratization and international organization membership
No full textABCDE: Appearance-Based Confidence Detection by Evaluating Gaze Behavior Using Deep Learning
No full text
The Emotional Microfoundations of Organizational Paradox: A Large Language Model Approach
No full textCo-Movement, Factors, and Forecasting: Data-Driven and Neural Network Advances in Financial Equity Market Analysis
No full textIn recent decades, data-driven and neural network methods have gained significant attention among academics and financial practitioners in equity market research. This thesis comprises four quantitative studies introducing innovative, data-driven approaches to measuring and predicting key financial variables in equity markets. First, we analyze global equity market co-movement over 25 years using a dynamic spatial model, focusing on major financial crises. Second, we employ neural networks to forecast downside deviation for investment factor timing and evaluate the results through an investment strategy. Third, we develop a neural network-based approach to predict daily realized volatility by transforming intraday data into images and using them as predictors. Finally, building on these insights, we propose a mixed-input, mixed-frequency neural network for next-day volatility forecasting, integrating intraday images, heteroscedastic autoregressive regressors, and market data. Given equity markets' increasing complexity and rapid evolution, this dissertation contributes to equity market research by advancing measurement and forecasting modeling techniques. This contribution is further reinforced by the publication of two of the four studies in international peer-reviewed journals: Dirkx and Heil (2022) in Expert Systems with Applications and Heil et al. (2022) in the Journal of International Money and Finance
