Dynamic mixture models for financial time series

Inaugural -Dissertation zur Erlangung des Grades eines Doktors der Wirtschafts -und Sozialwissenschaften der Wirtschafts -und Sozialwissenschaftlichen Fakultät der Christian -Albrechts -Universität zu Kiel The objective of this study is the development and application of models for financial time series to which normal mixture distributions are central. The use of mixed normal distributions for modeling the returns of financial assets is appealing because it maintains the assumption of conditionally normally distributed asset returns, yet can still adequately capture often observed "stylized facts" of both conditional and unconditional return distributions, in particular, fat-tailedness and asymmetries. The main part of this study is devoted to the development and investigation of univariate dynamic mixture models for financial time series, namely, mixed normal and Markov-switching GARCH models. In the second part of the study, we also consider multivariate problems, such as portfolio choice when the asset returns under study have a multivariate normal mixture distribution