Modeling the term structure of government bond yields with a differential evolution algorithm

Modeling the term structure of government bond yields is of major interest to macroeconomists and financial market practitioners. It is crucial for bonds and derivatives pricing, risk management, revealing market expectations, and essential for monetary policy decisions. This paper proposes the use of differential evolution to estimate the term structure of government bond yields considering parsimonious modeling. Differential evolution avoids non-convergence problems and the high instability of model parameter estimation caused by the need to choose initial values by the curve fitting algorithms of traditional optimization methods. Results show that differential evolution algorithm gives more accurate yield curves than nonlinear least squares and genetic algorithms, highlighting its potential as a tool for term structure of interest rates estimation.