A Time-Change Approach for Short-Term Interest Rate Modelling

The dynamics of interest rates have significantly evolved due to recent global economic events, such as the Covid-19 pandemic and ongoing economic uncertainty. These changes challenge traditional interest rate modeling approaches that rely on historical data and (log-)normality assumptions. Accurate interest rate risk prediction has become increasingly complex, affecting financial institutions across the board. This report delves into the current status of our research on interest rate modeling, presenting sophisticated and adaptable methodologies to manage these risks effectively.

Summary

• Challenges of Traditional Models: Recent global events have rendered traditional interest rate models based on historical data and (log-)normality assumptions less reliable.
• Impact on Financial Institutions: Banks, insurance companies and pension funds face heightened risks in asset and liability management due to inaccurate interest rate predictions.
• Transparency and Regulatory Scrutiny: Increased regulatory focus on the transparency and interpretability of interest rate risk models underscores the need for robust methodologies.
• Alternative Modeling Approaches: Our research explores the Subordinated Ornstein-Uhlenbeck (SubOU) process model, which shows promise as a more accurate alternative to the traditional shifted log-normal approach.
• Future Directions: Ongoing research aims to deepen our understanding of the SubOU model's practical implications for interest rate forecasting and risk management.

To gain comprehensive insights into our research and the proposed alternative modeling approaches, download the full report now.