Kripto Para Piyasasında Yüksek Frekanslı Verilerde Sürü Davranışı
Herding Behavior in High-Frequency Data in the Cryptocurrency Market

Summary/Abstract: This study aims to contribute to the literature by investigating whether cryptocurrency investors exhibit herding behavior. Moreover, this study examines herding behavior in the cryptocurrency market by using high-frequency data instead of daily frequencies. Given the global nature of the cryptocurrency market, intraday data facilitates a more precise analysis. In this context, the possibility of changing herding behavior with changing time intervals is investigated. In this way, we try to identify at which time intervals investors follow the movements of other investors without making sufficient use of their own information and evidence. This study investigates whether cryptocurrency investors exhibit herding behavior using both Cross-Sectional Standard Deviation (CSSD) and Cross-Sectional Absolute Deviation (CSAD) methodologies. The time dimension of herding behavior is also examined by conducting herding behavior analysis at different time intervals, including 15 minutes, 30 minutes, 1 hour, 2 hours, 3 hours, 6 hours, 12 hours, and daily data sets. The analysis covers the period from January 1, 2017, to December 31, 2022 and consists of major cryptocurrencies (BTC, ETH, LTC, XRP, and BCH) representing more than 70% of the total market capitalization. In this study, we find no evidence of herding behavior in high-frequency data, considering the entire period. Extreme price movements can be explained by rational asset pricing models. In other words, the trading decisions of cryptocurrency investors do not mimic the behavior of other investors, and investors make decisions based on their own information. Moreover, it can be argued that negative effects are more dominant than positive effects for cryptocurrency investors. As a result, cryptocurrency investors are more sensitive to negative information than positive information. The results of this study have several implications for the cryptocurrency market. First, the absence of herding behavior suggests that cryptocurrency investors are more likely to make rational investment decisions based on their own information. This is in contrast to traditional financial markets, where herding behavior is often observed. Second, the results show that cryptocurrency investors are more sensitive to negative information than positive information. This may be due to the fact that the cryptocurrency market is still a relatively new and volatile market. As a result, investors may be more likely to sell their positions in the face of negative news than to buy their positions in the face of positive news. Finally, the results of this study particularly emphasize the dominance of savvy investors and possibly institutional investors in the cryptocurrency market. This may be because these investors are more likely to have access to information and resources that enable them to make rational investment decisions. In addition, cryptocurrencies with significant market capitalization may not have been able to observe herding behavior because they are held by more rational investors. The absence of herding behavior in high-frequency data suggests that investors in the cryptocurrency market incorporate information into prices quickly, indicative of an efficient market in short time intervals. Our results suggest that investors may be less concerned about sudden and irrational market movements caused by mass behavior in short time intervals. Moreover, this suggests that investors are not overly influenced by the actions of others or in a way that can lead to rapid market movements. Hence, our study highlights the importance of the absence of herding behavior in high-frequency cryptocurrency data and provides valuable insights for efficient markets, investment strategies, and risk management practices.

• Details
• Contents