Time-frequency dynamics of public sentiment transmission in GCC equity markets: evidence from textual analysis

Purpose: This study aims to investigate the time-frequency dynamics of public sentiment transmission in Gulf Cooperation Council (GCC) equity markets, using a novel Twitter-based sentiment index for the GCC stock market. Design/methodology/approach: This study constructs a comprehensive Twitter-based sentiment index for the GCC equity market, using 1,107,449 tweets spanning the period 2014–2022. The FinBERT deep learning model is used to perform the sentiment analysis. Moreover, the authors use complete ensemble empirical mode decomposition with adaptive noise-based variable-lag transfer entropy to examine the information flow. Findings: The results reveal a significant negative association between the GCC Twitter sentiment index and stock returns in the GCC equity market. The transfer entropy result reveals significant information flow of public sentiment toward the GCC stock market. Moreover, time-frequency analysis indicates that sentiment transmission is more evident in the short- and medium-term scales. Research limitations/implications: The study’s findings have significant policy implications, indicating the need for a more complete governance mechanism for GCC stock markets to reduce the impact of irrational investor sentiment on stock returns. The study also emphasizes the need of taking social media sentiment into account when conducting market analysis and making decisions. Originality/value: This study adds to the existing literature by introducing a new Twitter-based sentiment index for the GCC stock market and using a novel methodology to investigate the time-frequency dynamics of sentiment transmission.