| Abstract [eng] |
The aim of the thesis is to investigate the dynamics between the price volatility of cryptocurrencies in order to determine whether there is a link between criminal activity and cryptocurrency price volatility. The thesis consists of three main parts. The first part covers the theoretical aspects of the use of cryptocurrencies for criminal activities, the types of criminal activities that take part in the cryptocurrency market and their impact on the price, an overview of the main price drivers of cryptocurrencies and the key differences between cryptocurrencies. In addition, this part involves the analysis of methods and variables used in related literature. The methodology section provides a detailed analysis of research models as well as selected variables, data period and formulated hypotheses. The practical part analyzes the results of the first, second and third empirical research. The third part represents the results of empirical analysis of dynamics between the price volatility of cryptocurrencies caused by criminal activity in cryptocurrency markets. After analysis of the literature, two statistical models were chosen to implement the three parts of the thesis. The first model (a Multivariate Generalized Autoregressive Conditional Heteroskedicity model) analyzes direct changes in cryptocurrency prices immediately after criminal incidents in the cryptocurrency market. In this model, the price of nine different cryptocurrencies was chosen as the dependent investigation variable and the independent variables were traditional financial instruments and the dates denoting the forty-four criminal incidents occurred in the cryptocurrency market. The second model (a Multivariate Generalized Autoregressive Conditional Heteroskedicity model) examines the effect of criminal incidents on the cryptocurrency market depending on the size of the loss. This model consists of the same variables as the first one, but in this case includes losses in US dollars during criminal incidents. The third research model (Dynamic Conditional Correlation Generalized Autoregressive Conditional Heteroskedasticity model) analyzes the changes in correlations between cryptocurrency markets caused by changing criminal activity incidents. All research models were developed by using software “OxMetrics” together with “Matlab” and “Eviews” The first research model revealed that the volatility of the cryptocurrency market is increasing due to criminal incidents in the cryptocurrency markets. Based on the results of the second research model, it was found that the volatility of cryptocurrency prices depends on the size of the losses caused by criminal activity. The third research model revealed that criminal incidents influence changes in the cross-correlations between cryptocurrency markets. The conclusions and suggestions summarize the results of the literature analysis and three research models. |
