Abstract

This thesis investigates the application of event study methodologies and cross-sectional factors in cryptocurrency markets, with a focus on understanding market dynamics and the drivers of cryptocurrency returns. Through two distinct but complementary studies, this work addresses both the methodological challenges of event studies in highly volatile markets and the role of novel factors in pricing ERC-20 tokens.
The first study examines the suitability of traditional event study methodologies in the context of cryptocurrency markets. Given the unique characteristics of cryptocurrencies—such as non-normal return distributions and extreme volatility—the study explores the efficacy of various parametric and non-parametric statistical tests. It identifies non-parametric approaches as more robust, particularly for smaller and highly volatile cryptocurrencies, and highlights the importance of sample size in achieving reliable results.
The second study investigates cross-sectional return predictors in the cryptocurrency market, with a specific focus on ERC-20 tokens. By leveraging both traditional factors such as size and momentum, as well as novel on-chain variables—including transaction value, transfer counts, and active addresses—the study constructs crypto-specific factors that provide deeper insights into token valuation and market behavior. It further demonstrates the relevance of these factors in explaining the variation in token returns.
Collectively, these studies contribute to the growing body of research on cryptocurrency markets by refining event study methodologies and introducing novel factors to better understand market reactions and return dynamics. The findings have broad implications for financial analysis in emerging and volatile asset classes, offering tools for researchers and investors to navigate the complexities of cryptocurrency markets.