Abstract
This dissertation explores the intricate relationship between AI-based cryptocurrencies and technology stocks, particularly focusing on their dynamic interactions during the COVID-19 pandemic. Using the Dynamic Conditional Correlation Generalized Autoregressive Conditional Heteroskedasticity (DCC-GARCH) model and various copula techniques, the study reveals that the correlations between these two asset classes are highly time-variant, with significant increases during periods of market turmoil. The heightened correlations observed during COVID-19 suggest a convergence in market behaviour, reducing the diversification benefits that investors typically seek when including these assets in their portfolios. In addition, the analysis of tail dependence using copulas- specifically the Clayton, Gumbel, Joe and tCopula- identifies substantial upper and lower tail dependencies, with the t-Copula providing the best fit for capturing symmetric extreme co-movements. These findings indicate that volatility spillovers between AI-based tokens and technology stocks amplify overall portfolio risk, especially during extreme market conditions. The implications of this research are twofold: for investors and asset risk managers; it underscores the necessity of dynamic portfolio management strategies that account for evolving dependencies and tail risks; and for policymakers and market regulators, it highlights the need for robust regulatory frameworks that can adapt to the unique risks posed by these emerging digital assets. By addressing a gap in the existing literature, this research provides valuable insights into the complex interplay between AI-based tokens and traditional financial markets, offering guidance for more effective risk management and investment decision-making in an increasingly interconnected financial landscape.