COMPSTAT 2022: Start Registration
View Submission - COMPSTAT2022
A0669
Title: Estimating crypto market betas Authors:  Jan Sila - UTIA AV CR, v.v.i. (Czech Republic) [presenting]
Ladislav Kristoufek - Institute of Information Theory and Automation, Czech Academy of Sciences (Czech Republic)
Abstract: The seminal financial problem of estimating market betas for crypto assets is investigated. We present empirical evidence of limited predictability of the future OLS betas, the theoretically optimal hedge against market risk. Our main finding is that non-OLS procedures better predict future realized OLS betas. This result correlates with recent stock market literature, but the general stability of crypto betas is much lower. Our work has implications for modern asset pricing theories measuring exposures to risk factors. The results also show a strong dependence on the selection of the market index, where we compare the ones recently used in the literature and the industry.