Herding Behavior in Cryptocurrency Markets: A Behavioral Perspective

Abstract

This study examines herding behavior in the cryptocurrency market by utilizing the Cross-Sectional Absolute Deviation (CSAD) and Cross-Sectional Standard Deviation (CSSD) models to assess the dispersion of returns. Additionally, Granger Causality tests were employed to investigate the association between market returns and herding behavior. The data for this analysis are sourced from major cryptocurrency exchanges, covering a range of cryptocurrencies including Binance Coin, Bitcoin, Solana, Ripple and Ethereum. The dataset includes monthly returns over time frame 2019 to 2023, to capture both bullish and bearish market periods. Data is collected from publicly available platforms such as CoinMarketCap, Binance API, and Yahoo Finance. The results indicate the presence of herding, particularly during periods of market stress or strong trends, with asymmetric herding observed between bullish and bearish market conditions. Specifically, herding behavior is more pronounced during bearish days compared to bullish days. However, the Granger causality tests reveal no significant causal relationship between market returns and herding behavior, suggesting that immediate price movements do not directly influence investor herding. This finding implies that other factors, such as market sentiment, investor psychology, and external market events, may play a more significant role in driving herding behavior in cryptocurrency markets. The study highlights the complexity of investor behavior in the cryptocurrency market and calls for further research into alternative drivers of herding, including sentiment analysis and the role of retail versus institutional investors