Herd behavior in stock markets during COVID’ 19 Pandemic: A machine learning approach

Abstract

The COVID-19 pandemic brought unprecedented volatility and uncertainty to global financial markets. During this period, the concept of herd behavior emerged as a prominent factor influencing stock market dynamics. Understanding and quantifying herd behavior patterns during the pandemic is crucial for predicting market trends, detecting potential bubbles, and improving risk management strategies. Herd behavior is characterized by a sudden mimicry among investors which causes temporary deviation of stock market prices. This deviation exacerbates in the presence of extreme conditions or events such as the recent pandemic. Based on social media contextual data this study aims to investigate the presence of herd behavior during the global COVID’19 pandemic. For this purpose state of the art machine learning algorithms are employed as opposed to traditional methodologies that are being used in the past literature for quantifying herd behavior. The surprising results reveal how role of media sentiment during the pandemic shaped the stock markets and investor behavior.