Investor comments on social media and index interaction: A study on the banking index

With the expansion of the internet and digital technologies, social media has transformed from a channel of personal interaction into a significant source of information in financial decision-making. This development aligns with behavioral finance literature, which highlights the psychological and irrational aspects of investor behavior beyond the assumptions of classical finance theories. In this study, 1,061,735 user comments posted between 2014 and 2023 on tr.investing.com, one of Turkey's leading financial platforms, were analyzed. The comments focused on nine banks listed in the BIST Banking Index: A dictionary-based sentiment analysis was performed using the NRC Emotion Lexicon, where positive and negative terms were identified and converted into time series for econometric modeling. Granger causality tests revealed that negative sentiment had a stronger effect on stock returns than positive sentiment, confirming the asymmetry widely documented in financial studies. Interestingly, shocks from positive comments were also found to have adverse effects on returns, a result interpreted within the framework of the "attention" hypothesis. In conclusion, the findings demonstrate that social media data can provide valuable insights into the interaction between investor sentiment and financial markets. The study contributes to approaches that move beyond rational models by incorporating behavioral dynamics and shows that lexicon-based sentiment analysis of social media comments can generate meaningful predictive signals for stock performance.