Beyond average: a study of herding effect in the brazilian stock market during Covid-19 pandemic using a quantile regression approach

Abstract

The research aimed to investigate the impact of the COVID-19 pandemic on the Brazilian stock market in terms of herd behavior. The analysis focused primarily on the first 100 days of the pandemic, and then in the first, second, and third waves. The chosen model for this analysis was the cross-sectional absolute deviation (CSAD) proposed by Chang, Cheng, and Khorana (2000). Unlike the cross-sectional standard deviation (CSSD) proposed by Christie and Huang (1995), CSAD is capable of detecting herd behavior under different market conditions, whether stressed or stable. The estimations were conducted using ordinary least squares (OLS) regression and quantile regression. While OLS regression estimates based on the mean of the distribution, quantile regression uses the median which is more robust to outliers and allows for analyzing various points along the distribution. The research was specified as documentary, descriptive, and quantitative. The sample consisted of 144 companies listed on B3 from January 2016 to September 2023. The results indicated that herd behavior was detected in the first 100 days of the pandemic, persisting until the end of the first wave. However, herd behavior was not evident in the second and third waves. It was also observed that the Brazilian market exhibited herd behavior in both bull and bear markets, with a tendency towards bull markets until the end of the first wave of the pandemic. This last result for the first wave was detected using quantile regression, not OLS. In the second and third waves, herd behavior was not evident under asymmetric conditions.