Robust Bond Investment Strategies under Parameter Uncertainty

Interest rate risk is a major concern for long-term investors. Standard models rely on a known distribution of this risk, while in practice the interest rate process is difficult to estimate precisely.

This dissertation consists of three chapters showing that long-term investors should not rely solely on a single set of parameter estimates, but should instead consider robust strategies that perform well when financial markets evolve differently than expected. The first chapter demonstrates that estimation errors in interest rate dynamics can strongly affect investment outcomes, sometimes making single-bond strategies more stable than two-bond strategies. The second chapter introduces parameter uncertainty and derives robust strategies
within a max–min framework, in which investors account for worst-case market conditions. The third chapter analyses uncertainty about a potential climate impact on interest rates. It studies the impact of estimation errors on utility and applies the max-min approach to derive robust strategies that account for these errors.