Essays on cryptocurrencies and investor attention

This thesis presents three chapters related to cryptocurrency and investor attention measures. Examining investor behaviour in the cryptocurrency market is important for understanding cryptocurrency market dynamics. The first chapter examines the impact of investor attention measures on cryptocurrency market returns between 2017 and 2022. Using the Google search volume index (GSVI) and Wikipedia (WIKI) searches for investor attention measures, we find no association between Bitcoin and Ethereum returns in up to eight periods and investor attention measures in the most recent period, 2020 to 2022. This new finding enriches the literature by revealing that the association between investor attention and Bitcoin and Ethereum returns has recently weakened. In the second chapter, we analyse the effect of the GSVI and WIKI investor attention measures on cryptocurrency return volatility. In particular, we build on the work of Figà-Talamanca et al. (2019), which studies the effect of GSVI on Bitcoin volatility, extending in terms of cryptocurrencies considered, sample period, and investor attention measures. Employing GARCH and EGARCH, we find that both GSVI and WIKI have a negative impact on the volatility of Ripple returns in the mean equation, also significantly increasing its volatility. Furthermore, these measures of investor attention significantly increase the volatility of all cryptocurrency returns. In the third and final chapter of the thesis, we study herding behaviour in the cryptocurrency market during bearish and bullish conditions by employing cross-sectional absolute deviation (CSAD) (Chang et al., 2000), and during extreme sentiment conditions based on GSVI and WIKI sentiment measures 2017 to 2024. Including WIKI as a sentiment proxy is a novel contribution to the extant literature on herding in the cryptocurrency market. In line with our previous chapters, we find significant results related to WIKI. We find herding behaviour in the cryptocurrency market and herding predominantly occurring during bearish markets. However, we find less evidence of herding during extreme positive days in the recent period.