According to this article, in the ISDA SIMM methodology, the delta sensitivity for interest rate products is calculated using

$$ \text{delta} = \frac{V (x + 1 \text{bp}) - V (x - 1 \text{bp})}{2} $$

However, when I look into the ISDA SIMM 2.5 documentation, I fail to find any reference to the formula above. Could you please opine what is the basis of this delta calculation?

  • 1
    $\begingroup$ please keep in mind that if the interest rates are high enough, like Argentina's, then in practice you may need to use a bigger bump than 1bp. $\endgroup$ Commented Sep 26, 2023 at 19:50
  • $\begingroup$ ISDA is sadistic. The 30-page document you linked to is just an appetizer. If you want the main course (which is some 99% of the documentation), you need to get your employer's lawyers to talk to ISDA and then they may grant you access to a server from which you can download the proper documentation, which is some 40 MB heavy. $\endgroup$ Commented Nov 25, 2023 at 20:29
  • $\begingroup$ It is just the Central Difference or Symmetric difference formula for approximating a derivative en.wikipedia.org/wiki/Symmetric_derivative $\endgroup$
    – nbbo2
    Commented Nov 26, 2023 at 15:38
  • $\begingroup$ @RodrigodeAzevedo thanks for your comment. While your suggestion may be ideal, but I don't think that will be practical. As a general researcher I can only have access to public information (paid or free) not insider information $\endgroup$
    – Bogaso
    Commented Nov 26, 2023 at 17:17
  • $\begingroup$ @Bogaso Are you affiliated with a university or research institute? If so, use your academic email address and write to [email protected]. I am under the impression that ISDA only cares about waiving any legal liability for your potential misuses of SIMM $\endgroup$ Commented Nov 26, 2023 at 17:19


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.