VAR: History or Simulation?

This paper assesses the performance of historical and Monte Carlo simulation in calculating VAR, using data from the Greek stock and bond market. Our contribution to the fixed income VaR literature is twofold in terms of the chosen interest rate process, and the method (Principal Components Analysis) that is used to calculate the interest rate volatilities. We find that while historical simulation results in over-commitment of capital for linear stock portfolios, the results for non-linear bond portfolios are less clear.