2013 Sept 5i

Discussion in 'CM2' started by Jia Syuen, Jul 12, 2021.

  1. Jia Syuen

    Jia Syuen Very Active Member

    May I know how we get the variance parameter for S_t/S_0 as 0.25t? Thank you in advance.
     
  2. Steve Hales

    Steve Hales ActEd Tutor Staff Member

    Hi
    The solution to the SDE is:

    \(S_t = S_0 e^{0.275t + 0.5B_t}\)

    Therefore:

    \(ln(S_t/S_0) = 0.275t + 0.5B_t\)

    From the definition of standard Brownian motion, \(B_t \sim N(0,t)\), therefore \(0.5B_t \sim N(0,0.25t)\) and:

    \(ln(S_t/S_0) \sim N( 0.275t ,0.25t)\)

    Steve
     
  3. Jonathan

    Jonathan Member

    Hi Steve

    Can you please explain how you get to 0.5Bt~N(0,0.25t) from Bt~N(0,t)?

    Can you also please explain how you get the mu parameter for the distribution of St?

    Thanks
     
  4. Steve Hales

    Steve Hales ActEd Tutor Staff Member

    Since Bt~N(0,t) then 0.5Bt will be normally distributed, we just need to know its mean and variance:
    • E[0.5Bt]=0.5*E[Bt]=0
    • Var(0.5Bt) = 0.5^2 * Var(Bt) = 0.25t
    Therefore 0.5Bt~N(0,0.25t).
    Hope that helps.
     
  5. Jonathan

    Jonathan Member

    Thanks Steve that is really helpful. I knew that I was missing something simple...

    Can you explain how, in this question, we get to S_t ~ logN(logS_0 + 0.275t, 0.25t)?

    I understand the variance part as you have explained above.

    Thanks
     
  6. Steve Hales

    Steve Hales ActEd Tutor Staff Member

    Hi
    If you're happy with \(ln(S_t/S_0) \sim N( 0.275t ,0.25t)\) from above, then we just need to note that:

    \(ln(S_t/S_0) = ln(S_t) - ln(S_0)\)

    Therefore we have:

    \(ln(S_t) - ln(S_0) \sim N( 0.275t ,0.25t)\)

    which implies that:

    \(ln(S_t) \sim N(ln(S_0)+ 0.275t ,0.25t)\)

    Let me know if that's not what you meant.
     

Share This Page