Empirical Analysis of Herd Behavior in Borsa Istanbul

Abstract

Behavioral finance is a field that has grown toward the end of 20th century as a reaction to the efficient market hypothesis. This new field studies the effect of investor psychology on financial decisions and explains stock market anomalies in financial markets. Herding is such an anomaly that is defined as mimicking others? decisions or market trend. The aim of the study is to detect whether there is herding or not in Borsa Istanbul. To test the existence of herding, stock returns traded on Borsa Istanbul and BIST 100 Index as market indicator are used. Data covers daily returns from 1988 to 2014 and intraday returns from 1995 to 2014. Firstly, herding is analyzed based on the methodology of cross-sectional dispersion of the stocks developed by Christie and Huang (1995) and Chang, Cheng and Khorana (2000). The results indicate that there is no herding for both up and down markets for daily and intraday intervals in Borsa Istanbul. However, tendency of herding is higher in up markets. To enhance and compare the results, the methodology based on the cross-sectional volatility of beta coefficients suggested by Hwang and Salmon (2004) is used. This methodology has provided evidence of herding in Borsa Istanbul. It is also observed that investors follow the market trend more in session two markets rather than session one markets. Thus, it is concluded that investors imitate the others more under normal market conditions rather than noisy market conditions. These results are consistent with the assumptions of Hwang and Salmon (2004).