Does information entropy drive stock returns? Evidence from Twitter sentiments during the Russia-Ukraine war

This paper employs transfer entropy and Markov regime-switching regression to examine the relationship between Twitter sentiments and stock returns during the Russia-Ukraine war. We explore the impact of both positive and negative Twitter sentiment proxies on stock returns from G20 nations between February 24 and December 21, 2022. We found that sentiments towards the leaders of Russia and Ukraine had a more significant impact on stock returns than overall sentiments regarding the war. Negative sentiments towards Vladimir Putin substantially impacted most sampled countries’ stock markets. This resulted in a detrimental impact on market performance in several nations. We further discovered that across various time periods, the direction of the relationship between Volodymyr Zelensky-positive sentiments and stock returns altered. This study underscores the importance of considering social media sentiment for policymakers and portfolio managers in safeguarding financial markets and managing portfolios during an economic slump.