On the Suboptimality of Single-Factor Exercise Strategies for Bermudan Swaptions
In this paper we examine the cost of using recalibrated single-factor models to determine the exercise strategy for Bermudan swaptions in a multi-factor world. We demonstrate that single-factor exercise strategies applied in a multi-factor world only give rise to economically insignificant losses. Furthermore, we find that the conditional model risk as defined in Longstaff, Santa-Clara amp;amp; Schwartz (2001), is statistically insignificant given the number of observations. Additional tests using the Primal-Dual algorithm of Andersen amp;amp; Broadie (2001) indicate that losses found in Longstaff et al. (2001) cannot as claimed be ascribed to the number of factors. Finally we find that for valuation of Bermudan swaptions with long exercise periods, the simple approach proposed in Andersen (2000) is outperformed by the Least Square Monte Carlo method of Longstaff amp;amp; Schwartz (2001) and, surprisingly, also by the exercise strategies from the single-factor models