Herding Behavior in High-Frequency Data in the Cryptocurrency Market

İbrahim Korkmaz Kahraman

Makale detayları

Herding Behavior in High-Frequency Data in the Cryptocurrency Market

Yazarlar:

İbrahim Korkmaz Kahraman

Yayıncı:

Hitit University

Ana Başlıklar:

Davranışsal Finans | Finans

Dergi:

Hitit Sosyal Bilimler Dergisi Vol.17 Issue 1, pp.54-69

Belge Tipi:

Makale

Yayın Tarihi:

4/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001315011

pISSN:

2757-7449

Anahtar kelimeler:

Özet:

 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.

 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.

 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.

 Bu çalışma, kripto para yatırımcılarının sürü davranışı sergileyip sergilemediğini araştırarak literatüre katkıda bulunmayı amaçlamaktadır. Ayrıca, bu çalışma kripto para piyasasındaki sürü davranışını günlük frekanslar yerine yüksek frekanslı veriler kullanarak incelemektedir. Kripto para piyasasının küresel yapısı göz önüne alındığında, gün içi veriler daha hassas bir analizi kolaylaştırmaktadır. Bu bağlamda, değişen zaman aralıkları ile sürü davranışının değişme olasılığı araştırılmaktadır. Bu şekilde, yatırımcıların hangi zaman aralıklarında kendi bilgi ve kanıtlarını yeterince kullanmadan diğer yatırımcıların hareketlerini takip ettikleri tespit edilmeye çalışılmaktadır. Bu çalışma, hem Yatay Kesit Standart Sapma CSSD hem de Yatay Kesit Mutlak Sapma CSAD metodolojilerini kullanarak kripto para yatırımcılarının sürü davranışı sergileyip sergilemediğini araştırmaktadır. Sürü davranışının zaman boyutu da 15 dakika, 30 dakika, 1 saat, 2 saat, 3 saat, 6 saat, 12 saat ve günlük veri setleri olmak üzere farklı zaman aralıklarında sürü davranışı analizi yapılarak incelenmiştir. Analiz, 1 Ocak 2017den 31 Aralık 2022ye kadar olan dönemi kapsamaktadır ve toplam piyasa değerinin 70sinden fazlasını temsil eden başlıca kripto para birimlerinden BTC, ETH, LTC, XRP ve BCH oluşmaktadır. Bu çalışmada, tüm dönem göz önünde bulundurulduğunda, yüksek frekanslı verilerde sürü davranışına dair herhangi bir kanıt bulunamamıştır. Aşırı fiyat hareketleri rasyonel varlık fiyatlandırma modelleri ile açıklanabilir. Başka bir deyişle, kripto para yatırımcılarının alım satım kararları diğer yatırımcıların davranışlarını taklit etmiyor ve yatırımcılar kendi bilgilerine dayanarak karar almaktadır. Dahası, kripto para yatırımcıları için negatif etkilerin pozitif etkilerden daha baskın olduğu söylenebilir. Sonuç olarak, kripto para yatırımcıları negatif bilgiye pozitif bilgiden daha duyarlıdır. Bu çalışmanın sonuçları kripto para piyasası için çeşitli çıkarımlara sahiptir. İlk olarak, sürü davranışının olmaması, kripto para yatırımcılarının kendi bilgilerine dayanarak rasyonel yatırım kararları alma olasılıklarının daha yüksek olduğunu göstermektedir. Bu, sürü davranışının sıklıkla gözlemlendiği geleneksel finans piyasalarının aksine bir durumdur. İkinci olarak, sonuçlar kripto para yatırımcılarının negatif bilgiye pozitif bilgiden daha duyarlı olduğunu göstermektedir. Bu durum, kripto para piyasasının hala nispeten yeni ve değişken bir piyasa olmasından kaynaklanıyor olabilir. Sonuç olarak, yatırımcıların olumsuz haberler karşısında pozisyonlarını satma olasılığı, olumlu haberler karşısında pozisyonlarını satın alma olasılığından daha yüksek olabilir. Son olarak, bu çalışmanın sonuçları özellikle bilgili yatırımcıların ve muhtemelen kurumsal yatırımcıların kripto para piyasasındaki hakimiyetini vurgulamaktadır. Bunun nedeni, bu yatırımcıların rasyonel yatırım kararları almalarını sağlayacak bilgi ve kaynaklara erişimlerinin daha yüksek olması olabilir. Buna ek olarak, önemli piyasa değerine sahip kripto para birimleri daha rasyonel yatırımcılar tarafından tutulduğu için sürü davranışını gözlemlemek mümkün olmamış olabilir. Yüksek frekanslı verilerde sürü davranışının görülmemesi, kripto para piyasasındaki yatırımcıların bilgiyi fiyatlara hızlı bir şekilde dahil ettiğini ve bunun da kısa zaman aralıklarında etkin bir piyasanın göstergesi olduğunu göstermektedir. Sonuçlarımız, yatırımcıların kısa zaman aralıklarında kitlesel davranışların neden olduğu ani ve irrasyonel piyasa hareketlerinden daha az endişe duyabileceğini göstermektedir. Dahası, bu durum yatırımcıların başkalarının eylemlerinden aşırı derecede veya hızlı piyasa hareketlerine yol açabilecek şekilde etkilenmediğini göstermektedir. Dolayısıyla çalışmamız, yüksek frekanslı kripto para verilerinde sürü davranışının olmamasının önemini vurgulamakta ve etkin piyasalar, yatırım stratejileri ve risk yönetimi uygulamaları için değerli bilgiler sağlamaktadır.

Makale detayları

Herding Behavior in High-Frequency Data in the Cryptocurrency Market

Yazarlar:

İbrahim Korkmaz Kahraman

Yayıncı:

Hitit University

Ana Başlıklar:

Davranışsal Finans | Finans

Dergi:

Hitit Sosyal Bilimler Dergisi Vol.17 Issue 1, pp.54-69

Belge Tipi:

Makale

Yayın Tarihi:

4/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001315011

pISSN:

2757-7449

Behavioral Finance

Herding Behavior

Bitcoin

Ethereum

CryptocurrencyDavranışsal Finans

Sürü Davranışı

Bitcoin

Ethereum

Kripto Para

Sayı

Makale detayları

Yazarlar: İbrahim Korkmaz Kahraman

Yayıncı: Hitit University

Ana Başlıklar: Davranışsal Finans | Finans

Dergi: Hitit Sosyal Bilimler Dergisi Vol.17 Issue 1, pp.54-69

Belge Tipi: Makale

Yayın Tarihi: 4/2024

Belge Dili: Türkçe

Document ID (Org No.): 2001315011

pISSN: 2757-7449

Sayı

Yazarlar:

İbrahim Korkmaz Kahraman

Yayıncı:

Hitit University

Ana Başlıklar:

Davranışsal Finans | Finans

Dergi:

Hitit Sosyal Bilimler Dergisi Vol.17 Issue 1, pp.54-69

Belge Tipi:

Makale

Yayın Tarihi:

4/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001315011

pISSN:

2757-7449