Mortgage-Backed Securities: Products, Structuring, and Analytical Techniques

MONTE CARLO AND OAS ANALYSIS

As explained in Chapter 10, option-adjusted spread (OAS) is a metric that attempts to show how much incremental spread is being earned over a benchmark forward curve, taking account of the cost of options embedded in the security. The methodology involved in generating Monte Carlo simulations and calculating option-adjusted spreads and durations was outlined in Chapters 10 and 11. Our discussion here is focused more on using OAS as a method of gauging relative value, with the proviso that OAS can be utilized by investors in a variety of ways. While it is tempting to view the bond with the greatest OAS as the “best” bond, this is often not the case; as with other metrics, OAS and its related methodologies are tools to be used in evaluating bonds. The output generated by a model must be considered in light of both the model’s predilections and the limitations of the methodology itself. In this section, we highlight some ways that the methodology can be used, both on its own merit and as an input to other forms of analysis.

As discussed in Chapter 10, the calculation of OAS using Monte Carlo simulations is a highly complex process. The generation of the underlying yield curve, the first step in the process of generating the analysis, is a complex undertaking in its own right; the additional complexity of integrating the model’s other components and inputs, such as volatility and prepayment functionality, suggests that the results obtained can be ...