With the escalating influence of social media and innovative social trading platforms on financial markets, understanding the behavioural patterns of retail investors becomes increasingly imperative. This study employs an Agent-Based Simulation approach (ABS) to scrutinise the retail investors’ behaviour in a social network. We built on the simulation model developed by Feng et al. (2012) [Feng, L.; Li, B.; Podobnik, B.; and Stanley, H.E. (2012). Linking agent-based models and stochastic models of financial markets. PNAS, 109 (22) 8388-8393], and proposed an innovative opinion clustering process. In particular, retail investors are divided into influencers and followers, where followers are further segregated into trend-chasers, contrarian traders, and noise traders. Trend-chasers buy after positive signals released by their focal influencers and sell after negative; contrarian traders behave conversely; and noise traders, acting independently of incoming signals, add an unpredictable element to the system. The results suggest that our model provides a closer reproduction of “fat” tails of the empirical financial market data compared to the baseline model developed by Feng et al. (2012). Our study provides a nuanced understanding of retail investors’ behaviour. It contributes to a more comprehensive portrayal of the complex financial decision-making in the digital age.