When implementing a code for the new model, even if it provides sensible price, it is still a good idea to compare it against some benchmarks, even in the special case of constant volatility Black-Scholes model. For example, I am interested in the price for American-style Call with strike 100 and expiry in 1 year, on stock that is 100 now, pays 3 eur fixed discrete deterministic dividend after 0.5 years, in the 0 interest rate environment. Are there any source with similar examples?

  • $\begingroup$ You might need to try a few different implementations such as finite difference, trinomial tree, or Longstaff Schwartz. $\endgroup$ – Gordon Aug 13 '15 at 13:16
  • $\begingroup$ @Gordon: thanks, of course I can build other models to validate the target one, but they may be incorrect as well. Chance is small of course, but it would be even smaller if there were data available with benchmark examples, which was used by many many people, so that we'd all know the prices there are correct. $\endgroup$ – Ulysses Aug 13 '15 at 14:17
  • $\begingroup$ I do not think there is such a publicly available benchmark. Some vendor system may have such a pricer. In every big financial institution, there is a model validation group, where they use either an internal benchmark or a third party vendor system to validate the model from the trading desk. $\endgroup$ – Gordon Aug 13 '15 at 14:35
  • $\begingroup$ What is your model?, SV?, SVJ? CGMY? please add link or reference $\endgroup$ – user16891 Aug 13 '15 at 19:23
  • $\begingroup$ Why not use QuantLib to get some estimates? Alternatively you can go low tech and copy the tests in QuantLib. That has the added benefit that they sometimes get the numbers from the papers that introduced a method. $\endgroup$ – Bob Jansen Aug 13 '15 at 20:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.