Decision-Making Theory in Analyzing Investor Behaviour in the Bond Market

Abstract

Aim: study the integration of economic variables and behavioural data to make bond price forecasting more accurate and understand market dynamics across economies. Methodology: media-based sentiment analysis, Bayesian forecasting, and time series modelling were used to determine bond price movements. Conclusions: The results show how behavioural and sentimental data influence bond price forecasts, especially in the context of emerging markets where sensitivity to investor sentiment is high. The findings show that the extended relationships with structured economic variables were more prominent for developed economies. It was demonstrated how sentiment analysis can be integrated into traditional economic models to improve forecasting accuracy when capturing volatility periods of. So, it adds to its usefulness for capturing market dynamics during volatility periods. Originality: The study offers a conceptual methodological framework by combining bond market analysis using structured and unstructured data. This improves the overall understanding of the role of sentiment in financial forecasting and extends applicability in different economic contexts to a broad discussion. Limitations of the Study: The use of publicly available sentiment data has some biases, and further improvement of the analysis tool is needed. This methodology can be extended to other financial instruments in further studies, and variables can be included to increase robustness. Practical Implications: The obtained data allows financial analysts and institutional investors to understand how to use sentiment analysis in bond market decision-making.