EXHIBIT 8: The Benchmark Results for the Callable Range Accrual
Swap
This exhibit presents the results for model comparison. We benchmark the
lattice model under different drift approximation methods with several
standard market approaches, e.g., the regression-based in the full LMM
and the HJM trinomial tree, for both accuracy and speed. The trade is
the same as the one in Exhibit 6. The grid space is =1/8 and the number
of nodes is S =200. The column ‘Dif from MC’ = 1 – (current row
price) / (price of MC in LMM). All computational times are denoted in
seconds on a computer with a 2.33 GHz Duo Core CPU.