(FEB NEWS) On Tuesday, March 11, 2025, at 10:00 AM WIB, a Guest Lecture was held at Miendrowo Hall, 4th Floor, Main Building of FEB UNAIR. This event featured Prof. Roman Matousek from Queen Mary University of London, UK, as the keynote speaker, with Martha Ranggi Primanthi, SE., MIDEC., Ph.D., as the moderator.
The event began with an introduction by the moderator, who highlighted the crucial role of social media in financial markets, particularly in influencing the movement of UK Gilt Yields. Prof. Roman Matousek then delivered his presentation, emphasizing several key points, including the increasing influence of social media, especially X (Twitter), on investor sentiment. He explained how social media accelerates the dissemination of economic and political information and can trigger market volatility, as seen in the UK mini-budget crisis of 2022, which caused a surge in bond yields.
In his presentation, Prof. Matousek also highlighted the impact of retail investors and market volatility, where online communities like Reddit can influence bond trading trends. Additionally, misinformation circulating on social media often leads to overreactions in the market before official clarifications are made. To understand these effects, machine learning and Natural Language Processing (NLP) are employed to analyze sentiment on X (Twitter), using AI models such as BERT and LSTM to predict UK Gilt Yield movements.
Furthermore, he explained how the Bank of England responds to bond yield volatility through monetary policy and how speculation on social media frequently affects market expectations regarding interest rates. The main challenges in using social media sentiment for market analysis include data noise, misinformation, and regulatory concerns. Moving forward, AI is expected to become increasingly sophisticated in market analysis, though regulations on social media influence in financial markets may also become stricter.
After the presentation, a discussion and Q&A session took place. Several participants asked questions regarding sentiment analysis methods, regulatory impacts on social media, and how investors can filter credible information from social platforms. The event concluded at 12:00 PM WIB with closing remarks from the moderator, expressing gratitude to the speaker and participants for their engagement. It is hoped that the insights shared will provide valuable perspectives on the relationship between social media and government bond markets.
