I was reading this great answer: What are the advantages/disadvantages of these approaches to deal with volatility surface?

And I have the following question:

How to show that the forward volatility smile of Local Volatility flattens while the one from Stochastic Volatility does not?

Could you please provide a mathematical demonstration of this fact?


This answer answers my question. But in the following equation from the answer

$C(t,S;T\to T+\theta,K) := E^Q[(\frac{S_{T+\theta}}{S_{T}}-K)_+] =: C_{BS}(S=1,\theta,K;\Sigma(t,S;T\to T+\theta,K))$

I don't understand why the payoff of the forward start option is not rather

$ E^Q[(S_{T+\theta}-S_{T})_+] $

and why it is set $S=1$

  • 1
    $\begingroup$ Does this answer help? $\endgroup$
    – AKdemy
    Commented Sep 16, 2021 at 20:34
  • $\begingroup$ @AKdemy thanks a lot! This helps a lot! But I still have an additional question, could you please have a look on the EDIT? $\endgroup$
    – Joanna
    Commented Sep 16, 2021 at 21:15
  • 1
    $\begingroup$ It's just a question of absolute forward start and relative forward start. Relative forward start are more appealing in general since they have nice homogeneity properties under log-space invariant diffusion models (like BS or most SV models). In other words, the last equality of the first equation you mention (especially the $S=1$ part and keeping the relative moneyness $K$) can only be written under a homogeneous model such as BS. $\endgroup$
    – Quantuple
    Commented Sep 17, 2021 at 9:29


Your Answer

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