• The delta risk correlation ${\rho }_{kl}$ is set at 99% between sensitivities $W{S}_k$ and $W{S}_t$ with the same bucket (e.g. currency), the same assigned vertex, but different curves.
• The delta risk correlation ${\rho }_{kl}$ between sensitivities $W{S}_k$ and $W{S}_t$, within the same bucket, with a different vertex is given by

  $$max\left[e^{-\alpha *\frac{T_k-T_l}{min\left[T_k;T_l\right]}};40\%\right]$$

  where $T_k$ is the vertex that relates to $W{S}_k$ and $\alpha$ set at 3%.

• Between two sensitivities $W{S}_k$ and $W{S}_l$ within the same bucket (i.e. the same currency), different vertexes and different curves, the correlation ${\rho }_{kl}$ is equal to the correlation parameter specified above, multiplied by 99.90%.

Excel

RegisterorLoginto see the example. The registration is free.