Empirical Asset Pricing of Cryptocurrency

<p dir="ltr">This thesis consists of three essays that provide a comprehensive exploration of the cryptocurrency market, addressing significant gaps in the literature through a systematic review and empirical investigations into the asset pricing process.</p><p dir="ltr">The first essay systematically reviews 2,098 cryptocurrency publications in finance, identifying three primary research streams: features of cryptocurrencies, market behaviour, and blockchain implications. It synthesises diverse findings, resolves contradictions, and maps future research directions. Additionally, it highlights two critical research topics that form the foundation for the subsequent essays.</p><p dir="ltr">The second essay examines the impact of economic shocks on cryptocurrency asset pricing. The results show that incorporating sensitivities to unexpected changes in economic variables, such as global stock market returns, financial stress, and inflation expectations, significantly enhances the explanatory power of asset pricing models in explaining both time-series returns and cross-sectional expected returns. Furthermore, the sensitivities to economic shocks yield substantial abnormal returns over the long run, suggesting the presence of economic risk premia within the cryptocurrency market. For example, cryptocurrencies with higher sensitivities to global stock market shocks and inflation expectations outperform their counterparts by average weekly returns of 0.48% and 0.49%, respectively. Similarly, cryptocurrencies more vulnerable to spikes in fear sentiment and financial stress deliver long-run outperformance of 0.47% and 0.44% per week, compared with the more resilient coins.</p><p dir="ltr">The third essay investigates the time variation of factor premia in the cryptocurrency market, focusing on six categories of long-short factors, including size, momentum, liquidity, volatility, psychological, and economic sensitivities. The essay explores the influence of behavioural finance variables, economic and financial indicators, and cryptocurrency market state variables on factor returns. The findings reveal substantial time variation in cryptocurrency factor returns, shaped by three pivotal mechanisms: behavioural finance-driven mispricing, investor behaviour amid different market states, and fundamental forces from economic and financial conditions. For example, heightened fear sentiment enhances the performance of larger and more liquid cryptocurrencies, while bullish market conditions amplify returns for lottery-like assets. This research extends traditional asset pricing models to the cryptocurrency market, offering novel insights into the predictability of factor premia and providing practical implications for investors navigating this dynamic and volatile market.</p>