RT Dissertation/Thesis T1 Essays on modelling state-dependent dynamics : applications to financial time series A1 Kuck,Konstantin WP 2019/12/16 AB This thesis explores state-dependence in the context of financial market dynamics and cross-market linkages. Time-varying behaviour of financial markets is widely observed and implies that their price dynamics are characterized by state-dependence with regard to changing economic conditions. From a statistical perspective, this means that the (inter-)dependencies of financial variables are non-linear and cannot be adequately described in the context of linear models. Using non-linear econometric models like quantile (auto)regression and Markov-switching models, this thesis focuses on the following issues: 1. Are the dynamics among crude oil prices stable or time-varying? Are the crude oil markets generally integrated or regionalized? Is there a leading benchmark price? 2. How are the volatility dynamics of crude oil and precious metals affected by the level of volatility? Are there differences between crude oil and precious metals? 3. How fast do investors react to negative shocks in the equity market? Do negative shocks in the equity market affect the volatility of gold and what are the implications for the role of gold as a safe haven? 4. What can be learned from intra-day data about temporal dependencies and information processing in the foreign exchange (FX) market? K1 Zeitabhängigkeit K1 Nichtlineare Dynamik K1 Volatilität K1 Rendite K1 Rohstoffmarkt K1 Aktienmarkt PP Hohenheim PB Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim UL http://opus.uni-hohenheim.de/volltexte/2019/1681